Wael Kotb Ata Thesis Full-Text PDF

Intelligent Control of Tracked
Vehicle Suspension
A thesis submitted to the University of Manchester for the degree of

Doctor of Philosophy
in the Faculty of Engineering and Physical Sciences
2014
Wael Galal Kotb Ata
School of Mechanical, Aerospace and Civil Engineering

Contents
CONTENTS1 .......................................................................................................................................1
A List of Figures .........................................................................................................................5

B List of Tables .........................................................................................................................15
C Nomenclature ........................................................................................................................19
D Abstract .................................................................................................................................21
E Declaration ............................................................................................................................22
F Copyright ...............................................................................................................................23
G Acknowledgement ................................................................................................................24
1 CHAPTER 1: INTRODUCTION .................................................................................................25
1.1 Background .........................................................................................................................25

1.2 Aims and objectives ............................................................................................................26
1.3 Contributions to knowledge ................................................................................................26
1.4 Thesis organization.............................................................................................................28
2 CHAPTER 2: LITERATURE REVIEW ......................................................................................30
2.1 Introduction .........................................................................................................................30

2.2 Conventional suspension system .......................................................................................30
2.3 Unconventional suspension system ...................................................................................33
2.3.1 Active suspension system ...........................................................................................33
2.3.2 Semi-active suspension system ..................................................................................33
2.4 Types of dampers employed In tracked vehicle suspension ..............................................33
2.4.1 Hydraulic dampers ......................................................................................................33
2.4.2 Hydro-gas dampers .....................................................................................................35
2.5 Smart fluids .........................................................................................................................35
2.5.1 Electrorheological fluids ..............................................................................................36
2.5.2 Magnetorheological fluids ............................................................................................36
2.5.3 Comparisons of ER and MR fluids ..............................................................................36
2.6 Magnetorheological fluid dampers......................................................................................37
2.7 Magnetorheological damper equivalent models .................................................................39
2.7.1 Parametric models ......................................................................................................39
2.7.2 Non-parametric models ...............................................................................................40
2.8 Semi-Active control of tracked vehicle suspension fitted with MR dampers ......................41
2.8.1 Magnetorheological damper controllers ......................................................................42
2.8.2 System controllers .......................................................................................................43
2.9 Summary ............................................................................................................................45
3 CHAPTER 3: EVALUATION OF TRACKED VEHICLE PASSIVE SUSPENSION

PERFORMANCE UNDER BUMP TERRAINS .................................................................................47
1
Word count: Main text 56,137 words, Total text: 66,220 words (Max limit 80, 000 words)
1
3.1 Introduction ......................................................................................................................... 47
3.2 Half model of passive suspension for tracked vehicle ....................................................... 47
3.3 Bump excitations ................................................................................................................ 49
3.4 Simulink model ................................................................................................................... 50
3.5 Computer model validation................................................................................................. 53
3.6 Simulation procedures........................................................................................................ 54
3.7 Parametric analysis of suspension performance under bump terrains .............................. 56
3.7.1 Effect of speed on dynamic response of tracked vehicles traversing a bump ............ 56
3.7.2 Influence of damper locations ..................................................................................... 62
3.7.3 Influence of number of dampers ................................................................................. 65
3.7.4 Influence of damping coefficient ................................................................................. 79
3.7.5 Influence of suspension stiffness ................................................................................ 88
3.7.6 Influence of wheel stiffness ......................................................................................... 97
3.8 Summary .......................................................................................................................... 102

PERFORMANCE UNDER PERIODIC TERRAINS ........................................................................ 107
4.1 Introduction ....................................................................................................................... 107

4.2 Sinusoidal excitations ....................................................................................................... 107
4.3 Parametric analysis of suspension performance under sinusoidal terrains ..................... 108
4.3.1 Influence of damper locations ................................................................................... 108
4.3.2 Influence of number of dampers ............................................................................... 110
4.3.3 Influence of damping coefficient ............................................................................... 125
4.3.4 Influence of suspension stiffness .............................................................................. 134
4.3.5 Influence of wheel stiffness ....................................................................................... 143
4.4 Summary .......................................................................................................................... 151
5 CHAPTER 5: EXPERIMENTAL IDENTIFICATION OF HYDRAULIC, HYDRO-GAS AND

MAGNETORHEOLOGICAL DAMPER CHARACTERISTICS ...................................................... 152
5.1 Introduction ....................................................................................................................... 152

5.2 Experimental test setup .................................................................................................... 152
5.3 Analysis of the measured characteristics of dampers ...................................................... 154
5.3.1 Analysis of the measured characteristics of hydraulic damper ................................. 154
5.3.2 Analysis of the measured characteristics of hydro-gas damper ............................... 156
5.3.3 Analysis of the measured characteristics of MR damper.......................................... 159
5.4 Restoring force surface (RFS) mapping of damper characteristics using Chebyshev
Polynomial Fits ............................................................................................................................ 161
5.4.1 Restoring force surface (RFS) technique ................................................................. 161
5.4.2 Chebyshev polynomial fitting technique.................................................................... 162
5.4.3 Comparison of RFS characteristics of hydraulic, hydro-gas and MR dampers at 2 Hz
164
165
2
5.5 Modified Bouc-Wen Model of MR Damper .......................................................................178
5.5.1 Dynamic model formulation .......................................................................................178
5.5.2 Effect of model parameters on the predicted response ............................................180
5.6 Summary ..........................................................................................................................184
6 CHAPTER 6: DESIGN, MANUFACTURE AND INSTRUMENTATION OF TRACKED

VEHICLE SUSPENSION TEST RIG ..............................................................................................185
6.1 Introduction .......................................................................................................................185

6.2 Design of test rig ...............................................................................................................185
6.3 Drive system .....................................................................................................................187
6.4 Vibration system setup .....................................................................................................188
6.4.1 Data acquisition system and control instrumentation ................................................190
6.5 Experimental procedures ..................................................................................................190
6.5.1 Input excitation ..........................................................................................................190
6.5.2 Signal processing methods .......................................................................................192
6.6 Controllers development ...................................................................................................192
6.6.1 Skyhook control .........................................................................................................192
6.6.2 Hybrid control ............................................................................................................192
6.6.3 MR damper controller ................................................................................................193
6.7 Semi-active control of test rig model ................................................................................193
6.8 Summary ..........................................................................................................................196
7 CHAPTER 7: SEMI-ACTIVE CONTROL OF TEST RIG SUSPENSION WITH

MAGNETORHEOLOGICAL DAMPERS ........................................................................................197
7.1 Introduction .......................................................................................................................197

7.2 System identification .........................................................................................................197
7.3 Evaluation of suspension performance with passive MR dampers ..................................198
7.3.1 Influence of MR damper locations on suspension performance ...............................199
7.3.2 Influence of number of MR dampers on suspension performance ...........................200
7.4 Semi-active control of test rig with one MR damper .........................................................201
7.4.1 Open Loop control of test rig using one MR damper ................................................202
7.4.2 Skyhook closed loop control of test rig using one MR damper .................................203
7.4.3 Hybrid closed loop control of test rig using one MR damper ....................................205
7.5 Semi-active closed loop control of test rig using two MR dampers ..................................208
7.6 Comparisons of open loop and closed loop semi-active control of test rig using one and
two MR dampers .........................................................................................................................209
7.7 Validation of test rig model ...............................................................................................210
7.7.1 Frictional effects in test rig .........................................................................................210
7.7.2 Validation of undamped suspension natural frequencies .........................................211
7.7.3 Validation of undamped suspension acceleration responses ...................................212
7.7.4 Validation of damped suspension with MR passive damper .....................................224
7.7.5 Skyhook control validation using one MR damper ....................................................227
7.7.6 Hybrid control validation using one MR damper .......................................................229
3
7.7.7 Validation of skyhook control using two MR dampers .............................................. 231
7.8 Summary .......................................................................................................................... 232
8 CHAPTER 8: SEMI-ACTIVE CONTROL OF TRACKED VEHICLE SUSPENSION WITH MR

DAMPERS ...................................................................................................................................... 234
8.1 Introduction ....................................................................................................................... 234

8.2 Semi-active control of tracked vehicle suspension with MR dampers ............................. 234
8.2.1 Dynamic model of large scale MR damper ............................................................... 234
8.2.2 Fuzzy-hybrid system control ..................................................................................... 235
8.3 Road excitation profiles .................................................................................................... 239
8.4 Semi-active control results ............................................................................................... 240
8.4.1 System responses under bump excitation ................................................................ 240
8.4.2 System responses under sinusoidal excitation ......................................................... 249
8.4.3 System responses under random excitation ............................................................. 259
8.5 Summary .......................................................................................................................... 268
9 CHAPTER 9: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK ........... 269
9.1 Conclusions ...................................................................................................................... 269

9.2 Recommendations for future work ................................................................................... 272
A REFERENCES ..................................................................................................................... 273
APPENDIX A: TEST RIG DESIGN SCHEMATICS ....................................................................... 279
APPENDIX B: COMPRESSION SPRING DETAILS ..................................................................... 294
4
A List of Figures
Figure ‎2.1 Different types of tracked vehicles suspension system ...................................................30
Figure ‎2.2 M113 tracked vehicle (a) schematic view, (b) suspension components [13] ..................32
Figure ‎2.3 Connection between torsion bar and road wheel in tracked vehicle[15] .........................32
Figure ‎2.4 Sectional view of hydraulic damper (a) mono-tube type (b) twin tube type [22] ..............34
Figure ‎2.5 Sectional view of the rotary-vane damper [21] ................................................................35
Figure ‎2.6 Hydro-gas suspension (a) Typical unit [25] and (b) sectional view[24] ...........................35
Figure ‎2.7 Magnetic field effect on MR fluid [33]...............................................................................36
Figure ‎2.8 Sectional view of MR Damper [22] ..................................................................................38
Figure ‎2.9 Mono-tube MR Damper [40] ............................................................................................38
Figure ‎2.10 Twin tube MR Damper [40] ............................................................................................39
Figure ‎2.11 Double-ended MR Damper [40] .....................................................................................39
Figure ‎2.12 Schematic diagrams for MR damper parametric models ..............................................40
Figure ‎2.13 Idealized skyhook control schemes [63] ........................................................................44
Figure ‎3.1 Half tracked vehicle suspension model ...........................................................................48
Figure ‎3.2 Typical bump road profiles at vehicle speed of 10 km/h..................................................50
Figure ‎3.3 Top-level diagram of the passive suspension model .......................................................51
Figure ‎3.4 Under mask block diagram for first suspension station ...................................................52
Figure ‎3.5 BA responses of suspension under shallow bump excitations at different speeds .........58
Figure ‎3.6 AA responses of suspension under shallow bump excitations at different speeds .........59
Figure ‎3.7 BA responses of suspension under sharp bump excitations at different speeds ............60
Figure ‎3.8 AA responses of suspension under sharp bump excitations at different speeds ............61
Figure ‎3.9 Bounce acceleration responses under shallow bump excitation .....................................67
Figure ‎3.10 Angular acceleration responses under shallow bump excitation ...................................68
Figure ‎3.11 Bounce acceleration responses under medium bump excitation ..................................72
Figure ‎3.12 Angular acceleration responses under medium bump excitation ..................................73
Figure ‎3.13 Bounce acceleration responses under sharp bump excitation ......................................74
5
Figure ‎3.14 Angular acceleration responses under sharp bump excitation ..................................... 75
Figure ‎3.15 Influence of varying damping coefficient on BA responses under shallow bump ......... 82
Figure ‎3.16 Influence of varying damping coefficient on AA responses under shallow bump ......... 83
Figure ‎3.17 Influence of varying damping coefficient on BA responses under medium bump ........ 84
Figure ‎3.18 Influence of varying damping coefficient on AA responses under medium bump ........ 85
Figure ‎3.19 Influence of varying damping coefficient on BA responses under sharp bump ............ 86
Figure ‎3.20 Influence of varying damping coefficient on AA responses under sharp bump ............ 87
Figure ‎3.21 Influence of suspension stiffness on BA responses under shallow bump .................... 90
Figure ‎3.22 Influence of suspension stiffness on AA responses under shallow bump .................... 91
Figure ‎3.23 Influence of suspension stiffness on BA responses under medium bump.................... 93
Figure ‎3.24 Influence of suspension stiffness on AA responses under medium bump.................... 94
Figure ‎3.25 Influence of suspension stiffness on BA responses under sharp bump ....................... 95
Figure ‎3.26 Influence of suspension stiffness on AA responses under sharp bump ....................... 96
Figure ‎3.27 Influence of wheel stiffness on BA responses under shallow bump ........................... 100
Figure ‎3.28 Influence of wheel stiffness on AA responses under shallow bump ........................... 101
Figure ‎3.29 Influence of wheel stiffness on BA responses under medium bump .......................... 103
Figure ‎3.30 Influence of wheel stiffness on AA responses under medium bump .......................... 104
Figure ‎3.31 Influence of wheel stiffness on BA responses under sharp bump .............................. 105
Figure ‎3.32 Influence of wheel stiffness on AA responses under sharp bump .............................. 106
Figure ‎4.1 Typical sinusoidal road profiles at speed of 10 km/h .................................................... 107
Figure ‎4.2 Bounce acceleration responses under shallow sinusoidal excitation ........................... 113
Figure ‎4.3 Angular acceleration responses under shallow sinusoidal excitation ........................... 114
Figure ‎4.4 Bounce acceleration responses under medium sinusoidal excitation ........................... 118
Figure ‎4.5 Angular acceleration responses under medium sinusoidal excitation .......................... 119
Figure ‎4.6 Bounce acceleration responses under sharp sinusoidal excitation .............................. 122
Figure ‎4.7 Angular acceleration responses under sharp sinusoidal excitation .............................. 123
‎ .8 Influence of varying damping coefficient on BA responses under shallow sinusoidal road

Figure 4
profile .................................................................................................................................... 127
6
‎ .9 Influence of varying damping coefficient on AA responses under shallow sinusoidal road
Figure 4
profile ....................................................................................................................................128
‎ .10 Influence of varying damping coefficient on BA responses under medium sinusoidal

Figure 4
road profile ............................................................................................................................129
‎ .11 Influence of varying damping coefficient on AA responses under medium sinusoidal

Figure 4
road profile ............................................................................................................................130
‎ .12 Influence of varying damping coefficient on BA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................131
‎ .13 Influence of varying damping coefficient on AA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................132
‎ .14 Influence of varying suspension stiffness on BA responses under shallow sinusoidal

Figure 4
road profile ............................................................................................................................135
‎ .15 Influence of varying suspension stiffness on AA responses under shallow sinusoidal

Figure 4
road profile ............................................................................................................................136
‎ .16 Influence of varying suspension stiffness on BA responses under medium sinusoidal

Figure 4
road profile ............................................................................................................................139
‎ .17 Influence of varying suspension stiffness on AA responses under medium sinusoidal

Figure 4
road profile ............................................................................................................................140
‎ .18 Influence of varying suspension stiffness on BA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................141
‎ .19 Influence of varying suspension stiffness on AA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................142
‎ .20 Influence of varying wheel stiffness on BA responses under shallow sinusoidal road
Figure 4
profile ....................................................................................................................................144
‎ .21 Influence of varying wheel stiffness on AA responses under shallow sinusoidal road
Figure 4
profile ....................................................................................................................................145
‎ .22 Influence of varying wheel stiffness on BA responses under medium sinusoidal road
Figure 4
profile ....................................................................................................................................147
‎ .23 Influence of varying wheel stiffness on AA responses under medium sinusoidal road
Figure 4
profile ....................................................................................................................................148
‎ .24 Influence of varying wheel stiffness on BA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................149
‎ .25 Influence of varying wheel stiffness on AA responses under sharp sinusoidal road
Figure 4
profile ....................................................................................................................................150
7
Figure ‎5.1 ESH testing machine with experimental setup .............................................................. 153
‎ .2 Displacement, velocity and force time history of hydraulic damper under sinusoidal
Figure 5
excitation 1Hz and (a) amplitude 2.5 mm (b) amplitude12 mm ............................................ 154
‎ .3 Force-displacement and force-velocity dependence of the hydraulic damper under

Figure 5
excitation frequency 1 Hz and dynamic amplitudes 2.5 mm to 12 mm ................................ 155

Figure 5

Figure 5
excitation frequency 11 Hz and dynamic amplitudes 2.5 mm to 12 mm .............................. 156
‎ .6 Force-velocity characteristics for hydraulic damper at 10 mm amplitude and different

Figure 5
excitation frequencies ........................................................................................................... 156
‎ .7 Displacement, velocity and force time history of hydro-gas damper under sinusoidal
Figure 5
excitation at 1 Hz and a- 2.5 mm b- 12 mm amplitudes ....................................................... 157
‎ .8 Force-displacement and force-velocity dependence of the hydro-gas damper under

Figure 5

Figure 5

Figure 5
excitation frequency 11 Hz and dynamic amplitudes 2.5 mm to 12 mm .............................. 158
‎ .11 Force-velocity characteristics for hydro-gas damper at 10 mm amplitude and different

Figure 5
excitation frequencies ........................................................................................................... 158
‎ .12 Displacement, velocity and force time histories of MR damper under sinusoidal
Figure 5
excitation at 1 Hz: (a) 2.5 mm at 0 Amp (b) 12 mm at 2 Amp .............................................. 159
‎ .13 Measured characteristics of MR damper at 1 Hz excitation frequency and with

Figure 5
amplitudes up to 12 mm and 1 Amp (a) force-displacement and (b) force-velocity diagram
.............................................................................................................................................. 160

Figure 5
.............................................................................................................................................. 160

Figure 5
amplitudes up to 12 mm and 1 Amp (a) force-displacement diagram and (b) force-velocity
diagram ................................................................................................................................. 160
‎ .16 Measured characteristic of MR damper under different excitation frequencies,

Figure 5
amplitude 10 mm and 1 A (a) force-displacement diagram and (b) force-velocity diagram . 161
8
Figure ‎5.17 Projected phase plane of hydraulic damper under excitation frequency of 2 Hz ........164
‎ .18 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 2

Figure 5
Hz excitation and amplitudes up to 12 mm; measured data (solid line) and grid forces
(circles) ..................................................................................................................................166
‎ .19 Validation of experimental test data and Chebyshev approximation under excitation
Figure 5
frequency 2 Hz and amplitude 5 mm (a) Force-Displacement and (b) Force-Velocity.........167
‎ .20 Comparison of restoring force surface characteristics of dampers under excitation

Figure 5
frequency 2 Hz, amplitude up to 12 mm ...............................................................................168
‎ .21 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 6

Figure 5
Hz excitation and amplitudes up to 12 mm; measured data (solid line) and grid forces
(circles) ..................................................................................................................................172
‎ .22 Validation of experimental tests and Chebyshev approximation under excitation

Figure 5
frequency 6 Hz and amplitude 10 mm (a) Force-Displacement and (b) Force-Velocity.......173
‎ .23 Comparison of restoring force surface characteristics of damper under excitation

Figure 5
frequency 6 Hz, amplitude up to 12 mm ...............................................................................174
Figure ‎5.24 Schematic view of the Modified Bouc-Wen model of MR damper ..............................179
Figure ‎5.25 Block diagram of the Modified Bouc-Wen Simulink model for MR damper .................179
‎ .26 Comparison between force-displacement and force-velocity characteristics predicted

Figure 5
by modified Bouc-Wen and measured characteristics under excitation of 3 Hz, 10 mm
amplitude and current of 1 Amp ............................................................................................181
‎ .27 Comparison between force-displacement and force-velocity characteristics predicted

Figure 5
by modified Bouc-Wen and measured characteristics under excitation of 6 Hz, 10 mm
amplitude and current of 1 Amp ............................................................................................181
‎ .28 comparison between force-displacement and force-velocity characteristics predicted by

Figure 5
modified Bouc-Wen using different sets of parameters and the measured characteristics
under excitation of 3Hz and 10 mm amplitude at 1 Amp ......................................................183
Figure ‎6.1 Front View of Seven DOF Test Rig (physical model) ....................................................185
Figure ‎6.2 Side View of Seven DOF Test Rig (physical model) .....................................................186
Figure ‎6.3 Test rig primary components .........................................................................................187
Figure ‎6.4 Rig main structure with rails and carriages ....................................................................188
Figure ‎6.5 Drive mechanism and its components ...........................................................................188
Figure ‎6.6 Electric geared motor .....................................................................................................188
Figure ‎6.7 Schematic of Test rig and Acquisition instruments ........................................................189
9
‎ .8 Schematic representation of test rig showing accelerometers mount to measure body
Figure 6
and wheel responses ............................................................................................................ 190
Figure ‎6.9 CompactRio system architecture; (i) photograph and (ii) schematic representation .... 191
Figure ‎6.10 Swept sinusoidal input excitation up to 10 Hz at amplitude of 2.5 mm ....................... 191
Figure ‎6.11 Semi-active control system of test rig suspension with MR damper ........................... 194
Figure ‎6.12 Skyhook controller Simulink model ............................................................................. 195
Figure ‎6.13 Hybrid controller Simulink model ................................................................................. 195
Figure ‎6.14 Heaviside Simulink model of the MR damper controller ............................................. 196
‎ .1 Bounce acceleration transmissibility of the undamped suspension under swept

Figure 7
sinusoidal excitation.............................................................................................................. 197
Figure ‎7.2 Nyquist plot of Bounce acceleration of undamped suspension system ........................ 198
‎ .3 Bounce and angular acceleration responses of suspension with different numbers of

Figure 7
passive MR dampers at [wh15, wh125 and all wheels] under swept sinusoidal excitation . 201
Figure ‎7.4 Bounce and angular acceleration for different damper currents ................................... 202
Figure ‎7.5 Sample time traces of skyhook control method execution ............................................ 203
‎ .6 Effect of gains of the skyhook control on BA and AA responses of the suspension with
Figure 7
MR damper under swept sinusoidal excitation ..................................................................... 204
‎ .7 Effect of gains of the skyhook control on wheel acceleration response of the suspension
Figure 7
with MR damper under swept sinusoidal excitation ............................................................. 205
‎ .8 Effect of gains of the hybrid control on BA, AA and wheel responses of the suspension
Figure 7
with MR damper under swept sinusoidal excitation (tau=0.5) .............................................. 206
‎ .9 Effect of weighting factor,

Figure 7 of the hybrid control on BA, AA and wheel responses of the
suspension with MR damper under swept sinusoidal excitation (G=1000 Ns/m) ................ 207
‎ .10 Skyhook control of BA and AA of the suspension with two MR dampers located at
Figure 7
wheels 1 and 5...................................................................................................................... 208
‎ .11 Hybrid control of BA and AA of the suspension with two MR dampers located at wheels
Figure 7
1 and 5 .................................................................................................................................. 209
‎ .12 predicted and measured input acceleration and wheel 1 acceleration of the undamped
Figure 7
suspension under swept sinusoidal excitation and amplitude of 2.5 mm ............................. 211
‎ .13 Predicted acceleration responses of undamped suspension under sinusoidal excitation

Figure 7
of 2 Hz and damping coefficient (C=0 Ns/m) ........................................................................ 214
10
Figure 7
of 2 Hz and damping coefficient (C=10 Ns/m) ......................................................................214

Figure 7

Figure 7

Figure 7

Figure 7
of 3.5 Hz and damping coefficient (C=0 Ns/m) .....................................................................216

Figure 7
of 3.5 Hz and damping coefficient (C=10 Ns/m) ...................................................................217

Figure 7

Figure 7

Figure 7

Figure 7
of 4.3 Hz and damping coefficient (C=0 Ns/m) .....................................................................219

Figure 7

Figure 7

Figure 7

Figure 7

Figure 7
of 6 Hz and damping coefficient (C=0 Ns/m) ........................................................................221

Figure 7

Figure 7
11
Figure 7
of 6 Hz and damping coefficient (C=60 Ns/m)...................................................................... 223

Figure 7
of 6 Hz and damping coefficient (C=80 Ns/m)...................................................................... 223
‎ .33 Predicted and measured body acceleration responses of undamped suspension under
Figure 7
swept sinusoidal excitation and amplitude of 2.5 mm .......................................................... 224
‎ .34 Measured and predicted BA and AA responses of the damped suspension system with
Figure 7
one passive MR damper at different locations under swept sinusoidal excitation ............... 226
‎ .35 Validation of Simulink model and controlled MR suspension rig using skyhook control
Figure 7
(G=100 Ns/m) under sinusoidal excitation of 3.5 Hz and C=0 Ns/m .................................... 228
‎ .36 Validation of Simulink model and controlled MR suspension rig using skyhook control
Figure 7
(G=100 Ns/m) under sinusoidal excitation of 3.5 Hz and C=80 Ns/m .................................. 228
Figure ‎7.37 Skyhook control command voltage applied to MR damper ......................................... 229
‎ .38 Validation of Simulink model and controlled MR suspension rig using hybrid control
Figure 7
(G=50 Ns/m and =0.5) under sinusoidal excitation of 3.5 Hz and C=0 Ns/m .................... 230
‎ .39 Validation of Simulink model and controlled MR suspension rig using hybrid control
Figure 7
(G=50 Ns/m and =0.5) under sinusoidal excitation of 3.5 Hz and C=150 Ns/m ................ 230
Figure ‎7.40 Hybrid control command voltage applied to MR damper ............................................ 231
‎ .41 Validation of computer model and test rig with two MR dampers with skyhook control
Figure 7
(G=2000 Ns/m) under excitation frequency of 6 Hz ............................................................. 232
Figure ‎7.42 Command voltage applied to MR damper located at wheel station 1 ......................... 232
Figure ‎8.1 Schematic drawing of fuzzy logic controller .................................................................. 235
Figure ‎8.2 First input membership function ‘Body velocity’ ............................................................ 236
Figure ‎8.3 Second input membership function ‘Wheel velocity’ ..................................................... 237
Figure ‎8.4 Output membership function ‘Weighting factor’ ............................................................ 237
Figure ‎8.5 Surface viewer of the Fuzzy logic control ...................................................................... 237
Figure ‎8.6 Block diagram of 7-DOF suspension model integrated with fuzzy-hybrid controller ..... 238
Figure ‎8.7 Fuzzy-hybrid controller Simulink model......................................................................... 238
Figure ‎8.8 Typical bump, sinusoidal and random road profiles imposed at first wheel station ...... 239
Figure ‎8.9 Time histories of system responses under bump excitation at different speeds .......... 242
12
‎ .10 Time histories of wheels bounce displacement under bump excitation; speed 10 km/h
Figure 8
..............................................................................................................................................243
Figure ‎8.11 Time histories of wheels displacement under bump excitation; speed 20 km/h .........244
Figure ‎8.12 Time histories of wheels displacement under bump excitation; speed 40 km/h .........245
‎ .13 Damping forces and command voltage of various control methods for MR damper
Figure 8
mounted on the first wheel under bump excitation; speed 10km/h ......................................246
‎ .14 Bounce acceleration responses of suspension systems with different control methods
Figure 8
under sinusoidal excitation at different speeds .....................................................................251
‎ .15 Angular acceleration responses of suspension systems with different control methods
Figure 8
under sinusoidal excitation at different speeds .....................................................................252
‎ .16 Wheel displacement responses of suspension systems with different control methods
Figure 8
under sinusoidal excitation at speed 10 km/h .......................................................................253
Figure 8
Figure 8
Figure 8
mounted on the first wheel under sinusoidal excitation; speed 10km/h ...............................256
‎ .20 Bounce acceleration responses of suspension systems with different control methods
Figure 8
under random excitation at different speeds .........................................................................260
‎ .21 Angular acceleration responses of suspension systems with different control methods
Figure 8
under random excitation at different speeds .........................................................................261
Figure 8
under random excitation at speed 10 km/h ...........................................................................262
Figure 8
Figure 8
Figure 8
mounted on the first wheel under random excitation; speed 10km/h ...................................265
Figure A.1 Front, top and side view of the test rig ..........................................................................280
Figure A.2 Test rig main frame .......................................................................................................291
Figure A.3 Drive mechanism components ......................................................................................291
13
Figure A.4 Excitation masses assembly ......................................................................................... 292
Figure A.5 Wheel mass assembly .................................................................................................. 293
Figure A.6 Body mass assembly .................................................................................................... 293
14
B List of Tables
Table ‎2.1 Properties of ER vs. MR fluids [10] ...................................................................................37
Table ‎3.1 Half model suspension parameters for tracked vehicle [82-84] ........................................49
Table ‎3.2 Description and parameters of selected road profiles ......................................................49
Table ‎3.3 Predicted natural frequencies of 7-DOF suspension for laden vehicle.............................54
Table ‎3.4 Predicted and measured undamped natural frequencies and associated modes ............54
Table ‎3.5 Proposed various suspension configurations ...................................................................55
Table ‎3.6 MOF values of suspension configurations S1 under different bump excitations ..............63
Table ‎3.9 MOF values of suspension configurations S4 and S5 under different bump excitations .64
Table ‎3.10 MOF for various suspension configurations under different bump excitations ...............64
Table ‎3.11 PTP and RMS values of bounce acceleration under shallow bump excitation ..............76
Table ‎3.12 PTP and RMS values of angular acceleration under shallow bump excitation ..............76
Table ‎3.13 PTP and RMS values of bounce acceleration under medium bump excitation ..............77
Table ‎3.14 PTP and RMS values of angular acceleration under medium bump excitation .............77
Table ‎3.15 PTP and RMS values of bounce acceleration under sharp bump excitation .................78
Table ‎3.16 PTP and RMS values of angular acceleration under sharp bump excitation .................78
Table ‎3.17 Influence of variation of damping on PTP and RMS values under shallow bump ..........81
Table ‎3.18 Influence of variation of damping on PTP and RMS values under medium bump .........81
Table ‎3.19 Influence of variation of damping on PTP and RMS values under sharp bump .............81
‎ .20 Influence of variation of suspension stiffness on PTP and RMS values under shallow
Table 3
bump at different speed ..........................................................................................................92
‎ .21 Influence of variation of suspension stiffness on PTP and RMS values under medium
Table 3
‎ .22 Influence of variation of suspension stiffness on PTP and RMS values under sharp
Table 3
‎ .23 Influence of variation of wheel stiffness on PTP and RMS values at different speeds at
Table 3
shallow bump ..........................................................................................................................97
15
Table 3
medium bump ......................................................................................................................... 98
Table 3
sharp bump ........................................................................................................................... 102
Table ‎4.1 Description and parameters of selected road profiles .................................................... 107
Table ‎4.2 MOF values of configurations S1 under different sinusoidal excitations ........................ 109
Table ‎4.5 MOF values of configurations S4 and S5 under different sinusoidal excitations ........... 110
Table ‎4.6 PTP and RMS values of BA under shallow sinusoidal excitation ................................... 115
Table ‎4.7 PTP and RMS values of AA under shallow sinusoidal excitation ................................... 115
Table ‎4.8 PTP and RMS values of BA under medium sinusoidal excitation .................................. 120
Table ‎4.9 PTP and RMS values of AA under medium sinusoidal excitation .................................. 120
Table ‎4.10 PTP and RMS values BA under sharp sinusoidal excitation ........................................ 124
Table ‎4.11 PTP and RMS values of AA under sharp sinusoidal excitation .................................... 124
Table ‎4.12 Effect of damping on PTP and RMS values under shallow sinusoidal excitation ........ 133
Table ‎4.13 Effect of damping on PTP and RMS values under medium sinusoidal excitation ....... 133
Table ‎4.14 Effect of damping on PTP and RMS values under sharp sinusoidal excitation ........... 133
‎ .15 Effect of suspension stiffness on PTP and RMS values under shallow sinusoidal
Table 4
excitation ............................................................................................................................... 134
‎ .16 Effect of suspension stiffness on PTP and RMS values under medium sinusoidal
Table 4
excitation ............................................................................................................................... 137
‎ .17 Effect of suspension stiffness on PTP and RMS values under sharp sinusoidal
Table 4
excitation ............................................................................................................................... 138
‎ .18 Influence of variation in wheel stiffness on PTP and RMS values under shallow
Table 4
sinusoidal excitations ............................................................................................................ 143
‎ .19 Influence of variation in wheel stiffness on PTP and RMS values under medium
Table 4
sinusoidal excitations ............................................................................................................ 146
‎ .20 Influence of variation in wheel stiffness on PTP and RMS values under sharp sinusoidal
Table 4
excitations ............................................................................................................................. 146
16
‎ .1 Calculated Chebyshev coefficients
Table 5 i for hydraulic damper at excitation frequency of 2
Hz and amplitudes up to 12 mm ...........................................................................................169

Table 5 i for hydro-gas damper at excitation frequency of 2

Table 5 i for MR damper at excitation frequency of 2 Hz,
amplitudes up to 12 mm and 1 Amp .....................................................................................171

Table 5 i for hydraulic damper at excitation frequency of 6

Table 5 i for hydro-gas damper at excitation frequency of 6

Table 5 i for MR damper at excitation frequency of 6 Hz,
amplitudes up to 12 mm and 2 Amp .....................................................................................177
Table ‎5.7 Parameters of MR damper [58] .......................................................................................180
Table ‎5.8 Sets of parameters for fitting Modified Bouc-Wen model to experimental data..............182
Table ‎6.1 Excitation signal parameters ...........................................................................................191
Table ‎6.2 Rig suspension parameters ............................................................................................195
Table ‎7.1 Estimated loss factor and damping ratios .......................................................................198
Table ‎7.2 MOF values for suspension with one MR damper ..........................................................199
Table ‎7.3 MOF values for suspension with two MR dampers ........................................................200
Table ‎7.4 MOF values for suspension with three MR dampers ......................................................200
Table ‎7.5 MOF values for suspension with four MR dampers ........................................................200
Table ‎7.6 Predicted undamped natural frequencies and associated modes ..................................212
Table ‎7.7 Measured and predicted hull natural frequencies ...........................................................212
Table ‎8.1 Parameters of the MR damper fitted to M113 suspension system [34, 47] ....................235
Table ‎8.2 Fuzzy control rules ..........................................................................................................238
Table ‎8.3 PTP values and improvement of body acceleration measures under bump road excitation
and various speeds ...............................................................................................................247
‎ .4 PTP values and improvement of wheel displacement measures under bump road
Table 8
excitation and 10 km/h ..........................................................................................................247
Table 8
excitation and 20 km/h ..........................................................................................................248
17
Table 8
excitation and 40 km/h .......................................................................................................... 248
‎ .7 RMS values and improvement of BA and AA measures under sinusoidal road excitation
Table 8
.............................................................................................................................................. 257
‎ .8 RMS values and improvement of wheel displacement measures under sinusoidal road
Table 8
Table 8
Table 8
‎ .11 RMS values and improvement of BA and AA measures under random road excitation
Table 8
.............................................................................................................................................. 266
‎ .12 RMS values and improvement of wheel displacement measures under random road
Table 8
Table 8
Table 8
Table B.1 Test rig compression spring details ................................................................................ 294
18
C Nomenclature
Notation
th
cbi Damping coefficient of i damper (Ns/m)
th
fa,i Actual damping force of i damper (N)
th
fd,i Desired damping force of i damper (N)
Gg Groundhook control gain (Ns/m)
Gs skyhook control gain (Ns/m)
h Height of bump and/or sinusoidal road profile (m)
2
Iy Body moment of inertia about lateral axis 'y' (kg.m )
K Stiffness matrix
th
kbi Equivalent stiffness of i torsion bar (N/m)
th
kwi Equivalent stiffness of i road wheel (N/m)
th
li Distance between body center of gravity and i road wheel center (m)
M Mass matrix
mb Body mass (kg)
th
mwi Mass of i road wheel (kg)
V, v Vehicle speed (km/h)
Vmax Maximum voltage applied to MR damper (volts)
w Width of bump road profile (m)
Wn White noise of random excitation
z Relative displacement of damper ends (m)
Relative velocity of damper ends (m/s)
zb Vertical displacement of body center of gravity (m)
Vertical velocity of body center of gravity (m/s)
2
Vertical acceleration of body center of gravity (m/s )
19
th
zri Vertical displacement of i road excitation (m)
Vertical velocity of road excitation (m/s)
th
zwi Vertical displacement of i road wheel (m)
th
Vertical velocity of i road wheel (m/s)
th 2
Vertical acceleration of i road wheel (m/s )
Greek Symbols
θ Pitch angle about lateral axis 'y' (rad)
θ Angular velocity about lateral axis 'y' (rad/s)
2
θ Angular acceleration about lateral axis 'y' (rad/s )
λ Wavelength of sinusoidal road (m)
-1
ρ Road roughness of random excitation (m )
2 2
σ Covariance of random road (mm )
σg Groundhook damping force module
σs Skyhook damping force module
Weighting factor of hybrid controller, time delay
20
D Abstract
Vibrations caused by rough road excitations influence tracked vehicle dynamic performance.
Good capabilities of such vehicles like high mobility, manoeuvrability and comfort are guaranteed
by optimal suspension systems. The suspension systems of tracked vehicles are exposed to
extreme operating conditions. This creates a conflict between ride comfort and handling that is
even greater than the conflict between ride comfort and handling for general road vehicles. Tracked
vehicles must be able to traverse not only rough roads but also smooth terrains. The challenges in
developing an optimized suspension system for tracked vehicles include the high and changeable
damping forces required for tracked vehicles crossing rough terrains. The use of active or semi-
active suspension systems overcomes the limitations inherent in the conventional passive
suspension. However, active suspension systems are expensive, complicated to design and have
high power demand. Thus, semi-active suspension systems have emerged as a good compromise
between active and passive suspension system. There is considerable current research on the
applications of magnetorheological (MR) fluid dampers for semi-active suspensions of executive
brand of some cars. However, there is very little research on semi-active devices for tracked
vehicle suspension. In fact, currently, there is no commercially available large scale MR dampers in
the market that produce the high damping force to suit such applications.
In response to these requirements, this research proposes a novel semi-active tracked
vehicle suspension system that uses MR dampers to improve the ride comfort and handling
characteristics of tracked vehicles. It also assesses the dynamics of the new suspension with
various semi-active control methods. This study is conducted in four phases. The first phase
provides a numerical investigation on the dynamic performance of a seven-degrees-of-freedom (7-
DOF) passive suspension model of the armour personnel carrier (APC) M113 tracked vehicle. The
numerical investigation considers the influence of variation of five suspension design parameters
on the vehicle dynamic performance. These parameters include number, locations of hydraulic
shock absorber, damping coefficient, suspension and wheel stiffnesses. The results indicate that
the optimal suspension performance is attained by using two or three dampers. The best locations
for these dampers are at the extreme road wheels i.e. the first, second and last road wheel
stations. Moreover, the vehicle performance is reduced when the damping coefficient is increased.
Additionally, low suspension stiffness offers better vehicle ride while high wheel stiffness degrades
the vehicle performance. These results identify the limitations inherent in the conventional passive
suspension.
For the second phase, the dynamic characteristics of the hydraulic, hydro-gas and MR
dampers are experimentally measured and fitted using the Chebyshev orthogonal functions to
produce the restoring force surfaces for each damper, which are compared. On one hand, the
restoring force surfaces of the hydraulic and hydro-gas dampers show fixed properties at specified
frequencies. On the other hand, the restoring force surfaces of the MR dampers show properties
that can be controlled at the same specified frequencies by the variation of the applied current
levels. Thus, the potential and the effectiveness of the controllable properties of MR dampers for
semi-active vibration control is demonstrated. Also, in this phase, the best set of parameters to use
in the modified Bouc-Wen model to characterise the MR dampers, has been derived. The third
phase of the project is also experimentally based. A new and novel test rig which represents the 7-
DOF scaled suspension model of the tracked vehicle is designed and fabricated. The primary
purpose of the test rig is to evaluate the performance of the proposed suspension with MR
dampers. Furthermore, experiments are conducted on the test rig to evaluate some semi-active
control methods and their effectiveness in reducing suspension vibration. The results show that the
use of two or three MR dampers at the extreme wheels offers optimal suspension performance.
This confirms the numerical results that are derived from the full scale passive suspension system
with hydraulic dampers. The experimental results also show that skyhook control and hybrid control
(which combines groundhook and skyhook controls) of the semi-active suspension are more
effective in reducing the road-induced vibration and improving the suspension dynamic behaviours.
Also, validations of the predicted responses of the semi-active scaled MR suspension model with
the measured responses have been presented.
The fourth and final phase provides a numerical simulation on the development and
evaluation of the semi-active control methods for a full scale tracked vehicle suspension with MR
dampers using the validated suspension model. Three semi-active control strategies are proposed.
The first two controllers are the skyhook and hybrid controls which provide better suspension
performance. In addition, the third controller, which is an intelligent fuzzy-hybrid control system, is
used to optimize the suspension performance. The results from this intelligent system are
compared with the two traditional control methods (skyhook and hybrid controls) under bump,
sinusoidal and random excitations. It is shown that the proposed controller can enhance
simultaneously the vehicle ride and handling characteristics.
21
E Declaration
No portion of the work referred to in the thesis has been submitted in support of an application for
another degree or qualification of this or any other University or other institute of learning.
22
F Copyright
The author of this thesis (including any appendices and/or schedules to this thesis) owns certain
copyright or related rights in it (the “ opyright”) and he has given The University of Manchester
certain rights to use such Copyright, including for administrative purposes.
Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be
made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and
regulations issued under it or, where appropriate, in accordance with licensing agreements which
the University has from time to time. This page must form part of any such copies made.
The ownership of certain Copyright, patents, designs, trademarks and other intellectual property
(the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example
graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by
the author and may be owned by third parties. Such Intellectual Property and Reproductions
cannot and must not be made available for use without the prior written permission of the owner(s)
of the relevant Intellectual Property and/or Reproductions.
Further information on the conditions under which disclosure, publication and

commercialisation of this thesis, the Copyright and any IP and/or Reproductions
described in it may take place is available in the University IP Policy (see
http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=487) in any relevant Thesis
restriction declarations deposited in the University Library, The University Library’s
regulations (see http://www.manchester.ac.uk/library/aboutus/regulations) and in The
University’s policy on presentation of Theses.
23
G Acknowledgement
Most importantly, I would like to give all my thanks to God.
I would like to express my sincere gratitude to my supervisor Dr. S.O.Oyadiji for his supervision,
guidance, support and continuous encouragement during this thesis. His enthusiasm and interest
in the work are hopeful. He is always there, ready and welcome to help.
Also, I am grateful to thank all the University members for their kindness and support. I wish to
thank Mr. Philip Oakes, Technical Coordinator, and all his members for manufacturing my test rig.
I wish to thank Mr. David Mortimer, technical team leader, for his help and support during the
experimental work.
I would like to express my thanks to my wife, Eman Abo Gharam and my children Ziad and Zeina,
for their love and patience throughout this thesis.
I wish to express my deepest grateful to my parents, brothers and sisters for their prayers, love and
encouragement during my study.
Finally, I wish to thank Egyptian Ministry of Defence and Military Technical College for giving
me this valuable opportunity to study abroad in a wonderful and lovely Greater Manchester City in
United Kingdom.
24
1 CHAPTER 1: INTRODUCTION
1.1 Background
Traditionally, the suspension system of a general road vehicle should have the ability to reduce the
body accelerations and minimize the suspension travel. These two tasks improve the vehicle ride
comfort and handling characteristics. In fact the ride comfort and handling performance [1-2] are
the most important characteristics in suspension performance evaluation. The main aim in
suspension design and vibration control strategy is to achieve improvement in these
characteristics. But, the conflict between these two aspects makes it harder to achieve [3].
Regarding the tracked vehicles, the suspension system is exposed to extreme operating
conditions. This creates a conflict between ride comfort and handling that is even greater than
general road vehicles. Tracked vehicles must be able to traverse not only rough roads but also
smooth terrains. The challenges in developing an optimized suspension system for tracked
vehicles include the high and changeable damping forces required for tracked vehicles crossing
rough terrains. The fixed characteristics i.e. quasi-linear damping inherent in conventional dampers
prevents the damper from generating high damping forces and from varying the damping according
to the operating conditions.
The use of active or semi-active suspension systems overcomes the limitations inherent in the
conventional passive suspension. The active suspension includes actuators and sensors that can
be remotely controlled to give a force or dissipate energy from the suspension [4]. Despite the
accuracy of the active suspension system, there is increasing concern that it has some
disadvantages. Owing to the high cost, complication in design and high power demand, the active
suspension is less popular and, therefore, has very low adoption rate in suspension design [3].
A compromise between active and passive suspension system is, a semi-active suspension
system that was proposed in [5-7]. In this type of suspension the damping force can be altered in
response to the outside conditions. One of the key advantages of the semi-active suspension is
that variation of the damping force does not consume high power and even need low power supply.
Another benefit of using semi-active suspension is explored in the event of control system failure.
In that case the semi-active suspension still behaves like a passive one. For these reasons, the
semi-active suspension combines some of the advantages of both the passive and the active
suspension.
In conventional semi-active suspension, altering the damping force is made by changing the orifice
area of the oil-filled damper. In this way the resistance to fluid flow is also altered. In practice, due
to the mechanical movement inherent in the operation, the change is very slow. A new class of
smart fluids based semi-active dampers or Magnetorheological (MR) dampers, have achieved a
wide application from the academia to the practical industrial world. This is caused by their
attractive characteristics and promising performance prospect to overcome limitations of the
existing oil-filled dampers in the market. There are two classes of the smart fluids which are the
electrorheological (ER) and magnetorheological (MR) fluids [8]. In consequence of their quick
response, usually milliseconds, and little power demands, ER and MR fluid suspensions have been
regarded as an attractive preference for realizing changeable damping [9]. These fluids can
25
instantly change their state from fluid to solid state when subjected to electric or magnetic field. For
the ER damper high voltage is needed, while the MR damper needs only low current and low
voltage, usually in the order of 1-2 Amps at 12-24 volts [10] for the available dampers in the
market.
For all these reasons MR dampers are considered valuable for realizing controllable high damping
force over a broad temperature range. Thus, MR dampers can be classified as ultimate devices for
improvement of the shock and vibration attenuation in the tracked vehicle suspension systems.
1.2 Aims and objectives

The scope of the current study is on the development of a novel semi-active tracked vehicle
suspension system and evaluation of its dynamic performance. Therefore, this research intends to
develop a semi-active tracked vehicle suspension that uses MR dampers to improve the ride
comfort and handling characteristics of tracked vehicles. Furthermore, the study aims to assess the
dynamics of the new suspension with various semi-active control methods under different
excitations. To achieve the research objectives, a series of targets are specified as follows:
1- Theoretical evaluation of the dynamic performance of a full scale tracked vehicle passive
suspension system.
2- Experimental identifications of the dynamic characteristics of hydraulic, hydro-gas and MR

dampers that identify the damping capabilities of each damper.
3- Design and fabrication of a new test rig representing a scaled model of a half tracked vehicle
suspension system with MR dampers.
4- Experimental evaluation of the dynamics of the scaled MR suspension test rig and
effectiveness of various semi-active control methods in reducing the road-induced vibration
that affect the suspension performance.
5- Development and evaluation of an intelligent control method for the semi-active suspension of
a full size tracked vehicle.
1.3 Contributions to knowledge

Novel semi-active tracked vehicle suspension systems incorporating MR dampers have been
developed to improve vehicle ride comfort and handling characteristics. Also, the performance of
the new suspension has been evaluated using various semi-active control methods. The results of
this study are expected to speed up the implementation of MR dampers in tracked vehicle
suspension systems. The specific contributions of the research can be listed as follows:
1- A computer simulation using 7 degrees of freedom (DOFs) of a full scale tracked vehicle
model has been developed and applied to the study of the dynamic performance of passive
suspension systems over different terrains. The model is validated using published
experimental data of the same tracked vehicle subjected to field tests. The suspension
behaviour is assessed using various design parameters such as number, location of damper,
damping coefficient, suspension and wheel stiffness. Various bump and sinusoidal road
26
profiles, which represent a wide range of working conditions and vehicle speeds, are used in
the simulation. The analysis reveals that the optimal suspension behaviour is attained by using
two or three hydraulic dampers. Furthermore, the best damper locations in the suspension are
at the first, second and last wheel stations. Also, low suspension stiffness offers better
dynamic ride while stiffer wheel stiffness reduces the vehicle performance.
2- The dynamic characteristics of the hydraulic, hydro-gas and MR dampers were experimentally
measured. The restoring force surfaces for each damper have been produced and compared
using validated Chebyshev polynomial approximation. The restoring force surfaces of the
hydraulic and hydro-gas dampers show fixed characteristics while the restoring force surfaces
of the MR dampers show controllable and highly nonlinear characteristics. A modified Bouc-
Wen model of the MR damper has been used and its predicted responses using different set
of parameters were fitted and compared to the measured characteristics. The best set of
parameters to use in the modified Bouc-Wen model to simulate the MR damper has been
identified.
3- A new and novel test rig for studying the dynamic performance of a scaled tracked vehicle
suspension system with MR dampers has been designed and fabricated. The optimal
suspension settings (best numbers and locations of the MR dampers) have been
experimentally identified under swept sinusoidal excitation. The results show that fitting the
suspension with two or three dampers at extreme wheel stations (wheels 1 and 5 or wheels 1,
2 and 5) offer the best suspension performance under different excitations. This confirms the
numerical results derived from the full scale passive suspension system with hydraulic
dampers.
4- Passive and semi-active control of the scaled MR suspension system have been employed
experimentally to show the effectiveness of the MR damper in semi-active vibration control. In
the passive control ”on state” mode, applications of different constant levels of current to the
MR damper produces corresponding constant levels of damping force such that at low
constant current levels, low levels of constant damping force are produced and at high
constant current levels, high levels of constant damping force are produced. This limits the
suspension capabilities and prevents the suspension from adapting to the operating
conditions. Traditional skyhook and hybrid controllers have been used as semi-active control
algorithms of the MR suspension test rig. The results of the semi-actively controlled
suspension with two MR dampers show a significant improvement in suspension performance.
In addition, a theoretical model of the controlled MR suspension test rig has been developed
and its predicted responses have been validated against the measured characteristics.
5- Simulation studies have been performed on a full size tracked vehicle suspension
incorporating three MR dampers using the validated suspension model with various semi-
active control methods. The simulation evaluates the semi-active suspension performance
under bump, sinusoidal and random road excitations. It also shows the effectiveness of
traditional and intelligent control methods in semi-active vibration control of the MR
27
suspension system. Furthermore, the results show that the semi-active control of the MR
suspension system has better dynamic performance than the performance of the conventional
passive suspension. Also, the use of the fuzzy-hybrid control as an intelligent system control
provides a significant improvement in both ride comfort and handling characteristics compared
to the traditional skyhook and hybrid controllers.
1.4 Thesis organization

This thesis includes 9 chapters. In chapter 1, the background, aims and objectives of the project
are discussed. The background includes a discussion of the conflicting requirements for vehicle
ride comfort and road handling. Also, the comparison between the active and semi-active
suspension is discussed. In chapter 2 a literature review of the previous published work relevant to
the research, addressed in the thesis, is presented. In the first section of this chapter a background
on tracked vehicle suspension system is highlighted. This includes a description of conventional
and unconventional suspension systems existing in tracked vehicles. This is followed by illustration
of different kinds of dampers employed in existing tracked vehicles such as the hydraulic and
hydro-gas dampers. Further details on the history of smart fluids including MR fluids are discussed.
Also, the chapter covers the basic information of the MR damper features and modelling taking
account of both parametric and non-parametric models. In addition, evaluation of MR damper
controllers and system controllers related to vibration control of suspension system are
summarized.
Chapters 3 and 4 present the evaluation of the dynamic performance of the full size tracked vehicle
suspension over bump and sinusoidal road profiles. This includes the influence of the variation in
the suspension design parameters such as number, location of dampers, damping coefficient,
suspension stiffness, and wheel stiffness. The bounce and angular accelerations of the body are
used as the suspension performance criteria to assess the vehicle performance under the
proposed excitations in the time and frequency domains.
Chapter 5 presents the experimental identification of the dynamic characteristics of hydraulic,

hydro-gas and MR dampers under different excitation frequencies, with variation of the dynamic
amplitudes and excitation current for the MR damper. The restoring force surfaces of the measured
dynamic properties of eachdamper are produced and compared using a validated Chebyshev
polynomial approximation. Also, a modified Bouc-Wen model of the MR damper is studied and its
predicted responses using different set of parameters are presented and compared to the
measured MR responses.
In Chapter 6, the design and fabrication procedures of a new and novel scaled test rig representing
tracked vehicle MR suspension system is presented. In addition, the vibration system setup,
experimental procedures and semi-active controller are presented. A Simulink model representing
the semi-active MR suspension of the test rig is developed. Chapter 7 presents the measured
suspension responses with passive and semi-active MR damper control. The influence of the
number and location of the MR damper on the suspension performance is investigated and the
optimal MR suspension settings are identified. This is followed by an investigation of the
effectivness of skyhook and hybrid control methods in controlling the suspension with one and two
28
MR dampers.Finally, a theoretical model of the semi-active suspension test rig is developed and
validated under sinusoidal excitation.
Chapter 8 is on the numerical simulation of the semi-active vibration control of a full size tracked
vehicle suspension with three MR dampers using a validated suspension model. The effectiveness
of the traditional (skyhook and hybrid) and the intelligent (fuzzy-hybrid) control methods in semi-
active vibration control of MR suspension are presented. The performance of the MR passive and
semi-active suspension are compared to the performance of the conventional passive suspension
under bump, sinusoidal and random excitations. Finally, chapter 9 presents the conclusions drawn
from this study and explores some ideas for the future applications of these results.
29
2 CHAPTER 2: LITERATURE REVIEW
2.1 Introduction
The most challenging objectives in upgrading a modern tracked vehicle’s capabilities are to
increase its dynamic performance, and to increase its maximum speed in particular when crossing
off roads. The achievement of vast stability for such vehicle, mainly when firing during movement,
is an issue. The ascending potential depends not only on the uneven terrain that the tracked
vehicle crosses and on its speed, but also on the tracked vehicle suspension system. Therefore,
the suspension system has a significant contribution in the improvement of tracked vehicle
dynamics. A review of tracked vehicles suspension system is presented accordingly.
2.2 Conventional suspension system

A tracked vehicle’s suspension incorporates all elements that connect the road wheels to the
vehicle hull (body) [11]. The tracked vehicle suspension system is classified into two types
according to the wheel connection to the hull. The first type is the block suspension and the second
one is the individual suspension as shown in Figure 2.1. In the block suspension type, two or more
wheels are linked together in one group and this group is connected to the hull. In the individual
suspension, every wheel is attached to the hull independently. The individual suspension system is
the most common suspension system currently used in the tracked vehicles.
(b) Individual suspension

(a) Block suspension
Figure ‎2.1 Different types of tracked vehicles suspension system
To obtain a close insight of a tracked vehicle suspension system, a full explanation of an Armoured
Personnel Carrier (APC) M113 undercarriage, which is shown in Figure 2.2(a, b) as an example, is
explained. Such a system consists of the following components:
 Road wheels: five on each side to support the vehicle weight.

 Road wheel arms: five on each side splined to individual torsion bars to suspends the vehicle
(Figure 2.2(b).
 Torsion bars: five on each side, anchored to the vehicle body to keep wheels on ground
(Figure 2.2(b).
 Track adjusters: secured to idler wheels to maintain track tension
30
 Shock absorbers: three per each side at first, second and fifth road wheels
Figure 2.3 shows the road wheel, wheel arm and torsion bar connection used in tracked vehicle.
The torsion bars are mounted inside the vehicle and extended across the width of the body. one
end of a torsion bar is splined to road wheel arm. The other end of the torsion bar is secured to the
opposite side of the vehicle body. the road wheels are mounted to crank-shaped arm (wheel arm)
which is connected to the body through bearing mounting.
The elastic elements (torsion bars) and shock absorbers, in tracked vehicle suspension play an
important role affecting essential mobility requirements of these vehicles. A torsion bar softens the
shocks generated when a tracked vehicle is driven over rough roads or uneven terrains in order to
minimize hull vibration. The shock absorber is responsible for damping the transmitted vibration
from road irregularities. The performance of the vehicle suspension system can be assessed with
regards to ride comfort, handling, and vehicle stability. Ride comfort is influenced by the absolute
acceleration of the vehicle body, whilst the handling is related to the relative displacement between
the vehicle body and the tires [12]. In contrast, vehicle stability relies on the continuous contact of
the tire on the ground. In fact, enhancement of the suspension system requires the improvement of
these three goals; however this perfection is an enormous challenge as these goals are in conflict
with each other [13]. In practice, if the ride comfort is an issue, then the damping needs to be low in
order that the damper may absorb the road excitation. This happens at the expense of the vehicle
stability which requires high damping i.e. a stiff damper.
Such suspension systems cannot carry out smooth movement and high troop’s fighting efficiency.
Thus the passive suspension limits the high mobility of such vehicles, as it only has the ability to
absorb external energy but not to adapt to the vehicle working conditions and excitations [14]. In
addition, there is a further limitation for the enhancement of tracked vehicles cross-country ability
and indeed passive suspension will be inadequate for future battles. Therefore, there is the need
for research investigations to be carried out for the enhancement of tracked vehicles performance.
There are two major classifications that introduce enormous upgrading in tracked vehicle mobility
which are, active and semi-active suspension system. The following paragraphs will discuss these
two types of suspension.
31
(a)
(b)
Figure ‎2.2 M113 tracked vehicle (a) schematic view, (b) suspension components [13]
Figure ‎2.3 Connection between torsion bar and road wheel in tracked vehicle[15]
32
2.3 Unconventional suspension system
As indicated earlier the classical (passive) suspension performance exploits fixed characteristics
for both spring and damper components, but the unconventional suspensions employ controllable
dampers, which offer accurate control strategy. There are two categories of the unconventional
suspension systems which are the active and semi-active suspension systems. The distinction
between these two classes are considered hereafter.
2.3.1 Active suspension system

This type utilizes force actuator, as an alternative of passive damper, to generate the required
damping force. The actuator can apply an independent force on the suspension to achieve the
crucial performance. This kind of active suspension can offer accurate control policy, which gives
enhanced settlement of ride quality and vehicle stability [16]. Concerning the crucial aspects of
tracked vehicles, all high mobility, fire control accuracy and ride comfort of the crew are assured by
superior suspension characteristics. Theoretically, fully active suspension provides accurate control
and high performance, but there are various commercial problems which need to be solved. These
difficulties include the high power consumption, requirements for lots of sensors, servo-valves and
sophisticated control algorithms [14]. Hence utilizing such suspension in real field conditions is
unfavourable.
2.3.2 Semi-active suspension system

In this kind of suspension, the conventional spring element is retained, but the damper is replaced
with a controllable one. The semi-active suspension system directs the damper, using an
embedded controller, to alter the damping level. The degree of damping force is established by the
controller based on a built-in control strategy. The generated damping force can either be in two
levels, namely minimum and maximum damping levels, or it can be altered in a range of damping
levels bordered by the minimum and maximum damping levels. The benefits of the semi-active
suspension are the simplicity of construction, lower power consumption and high reliability [12, 17].
Most of the advantages of a fully active suspension can be obtained with a semi-active suspension
system but without difficulty and high costs intrinsic to an active suspension system [18]. The semi-
active suspension system is better suited for achieving the necessary capabilities of tracked
vehicles, such as better and accurate performance [19]. In addition, the semi-active suspension
uses controllable dampers to obtain the required damping force. Such type of dampers can be of
the type ’’electrorheological’’ (ER) or ’’magnetorheological’’ (MR) dampers that are also called
smart devices and use smart fluids instead of hydraulic fluids.
2.4 Types of dampers employed In tracked vehicle suspension

This section starts with the discussion of the different types of shock absorbers employed in the
existing tracked vehicles. This includes the principle of operation and the components of each
shock absorber.
2.4.1 Hydraulic dampers

Most of the tracked vehicles utilize hydraulic dampers which are called hydraulic shock absorbers.
All the hydraulic shock absorbers work by the principle of converting the mechanical energy to
33
thermal energy. In this study this device is referred to as a damper, as this is more convenient
according to its applications. There are two types of these dampers which are the telescopic and
the rotary-vane types. The former has two similar configurations which are the mono-tube [20] and
twin dampers as shown in Figure 2.4, and the latter has rotating vanes [21] as shown in Figure 2.5.
The telescopic dampers in mono-tube type dampers consist of a piston which slides inside a
cylinder filled with hydraulic oil as shown in Figure 2.4(a). The piston divides the cylinder into two
chambers. There is also a floating piston that separates the oil and the gas which is usually
nitrogen. The damper has two strokes, compression and extension due to the inward and outward
movement of the piston. In the case of the twin tube damper, there is an outer tube which is called
a reservoir that encloses an inner cylinder as can be seen from Figure 2.4(b). In both types, the
piston has special outlet valves (orifices), not shown in the figure, which allows the oil to pass
through from the high pressure to the low pressure chambers. In the mono-tube damper the oil is
separated from the pressurized gas (nitrogen) by a floating piston. The gas allows enough room for
the oil to be displaced when the piston moves inwards (inside the cylinder) during the compression
stroke. In the extension stroke (rebound) the gas pushes the floating piston to move outwards after
the oil is transferred to the chamber with lower pressure.
The rotary-vane type is shown in Figure 2.5. The fluid-filled opposing chambers are interconnected
through a metering valve. Rotational force applied to the end of the wing shaft displaces the fluid
from the high-pressure chamber to the opposing low-pressure chamber. This causes a resistance
to relative motion and damping of the input force. The amount of damping torque is governed by
angular velocity of the load applied to the shock absorber and by the valves settings.
(a)
(b)
Figure ‎2.4 Sectional view of hydraulic damper (a) mono-tube type (b) twin tube type [22]
34
Figure ‎2.5 Sectional view of the rotary-vane damper [21]
2.4.2 Hydro-gas dampers

The hydro-gas unit [23] is a different type of suspension used in some tracked vehicles like the
Challenger battle tank [24]. The tank suspension has six independent hydro-gas units per each
side which offer vibration suppression. The hydro-gas unit includes the hydraulic cylinder with two
separated chambers and two pistons as shown in Figure 2.6. The two chambers are coaxial and
filled by oil and nitrogen gas. A floating piston separates the two chambers while the other piston is
connected to a crank via a connecting rod. A disc spring type damping valve is mounted between
the two pistons. The crank is integrated with the wheel axle arm pivot. The design principle of
hydro-gas unit is as the road wheel moves up, the crank rotates and moves the piston via the
connecting rod; the piston displaces oil through the damper valve and moves the separator piston.
This causes the gas to compress which produces the spring force. Most springs have a linear
spring rate, for example torsion bar, coil springs and rubber. Gas has a progressive rate which
increases depending on how fast it is compressed. A gas spring suspension will, therefore, result in
a softer rate around the static position (a soft ride over normal terrain) and a stiffer rate near full
deflection.
(a) (b)
Figure ‎2.6 Hydro-gas suspension (a) Typical unit [25] and (b) sectional view[24]
2.5 Smart fluids

In this section, a background overview of two types of the smart fluids is reported. These include
principle of operation and construction of both Electrorheological (ER) and Magnetorheological
(MR) fluids [8-9, 14]. Additionally, a comparison of ER and MR characteristics is discussed.
35
2.5.1 Electrorheological fluids
ER fluids usually consist of fine polarisable particles suspended in an insulated fluid like silicone or
mineral oil [26-27]. Without applying electric field, ER fluid continues to flow like Newtonian fluid
however, in the presence of an electric field, ER particles are joined together to form chains in the
direction of electric field. The aligned chains restrict the fluid movement. Consequently, a yield
stress is built up in the fluid [28]. It is obvious that the change in apparent viscosity of that fluid is
dependent on the change of the electric field. The significant effect of varying electric field appears
in the shear stress developed inside the fluid [29]. This enhancement in stress level takes place
rapidly with response of milliseconds. Hence, the resistance to motion of the fluid is restricted by
controlling of the electric field [30]
2.5.2 Magnetorheological fluids

Magnetorheological fluids are a class of smart fluids that have the ability to alter their nature or to
transform from liquid to solid state rapidly when subjected to an electric or magnetic field [9, 26].
These controllable fluids offer the innovative prospective of rapidly, and adaptively controllable
dampers, that have successful exploitation in the field of vibration control [31-32]. The rheological
properties of such fluids can vary according to the change in magnetic field, and attain their
greatest value when the applied magnetic field is normal to the flow of the MR fluid. Such MR fluids
possess the unique ability to be totally reversible and major changes in their strength are due to the
magnetic field as revealed in Figure 2.7. MR fluids consist of ferromagnetic particles suspended in
a carrier oil. The ferromagnetic particles are usually inexpensive particles such as carbonyl. A
typical fluid, like synthetic oil and silicone oil can be used as a carrier for MR fluids. Over time the
ferromagnetic particles tend to settle out of the carrier fluid due to the intrinsic density variation
between the particles and the fluid. One of the important principles, when making MR fluids, is to
add some chemicals to prevent sedimentation of the ferromagnetic particles and also to encourage
spreading of particles [8, 10].
Figure 2.7 Magnetic field effect on MR fluid [33]
2.5.3 Comparisons of ER and MR fluids

MR fluids have attained the phase of commercial applications faster than electrorheological (ER)
fluids. Various properties of distinctive ER and MR fluids are summarised in Table 2.1. Revealing
the same reaction time and plastic viscosity for both ER and MR fluids, MR fluids are insensitive to
contaminations and impurities, like water [10]. The negative aspects of ER fluid comprise high
sensitivity with temperature variations and slightly alteration in rheological properties. MR fluids
36
have a broad temperature range, in distinction to ER fluids. Mostly the operational temperature of
o
MR fluids simply span from – 40 to +150 C, regarding to ER fluids which are typically operate at
o
range +10 to +90 C. Furthermore, the operational voltage of MR fluids is lower than ER
analogous. MR fluid technologies put forward many benefits compared to conventional electro-
mechanical systems [13]. The quick response time usually less than 10 milliseconds, permits a
dramatic improvement in both ride comfort and stability. In addition, high efficacy in road shocks
and vibration isolation due to continuously variable control of damping force. As well as, its
simplicity in design as it has fewer moving parts approximately 60% less than valve based
dampers. Also, it has the highest dissipation of energy in contrast to other controllable devices.
All prior advantages take the MR technology from the laboratory field to the feasible commercial
applications. These various applications embrace both automotive and military vehicles
suspensions [13]. In automotive suspension, MR fluid replaces the usual hydraulic fluid in shock
absorber. The proven MR damper eliminates the compromise between ride comfort and safety that
correlated to the passive suspension systems [34-35]. For the military vehicle suspension, MR
dampers are a great innovation that gives more benefits and capabilities to the suspension system.
Semi-active dampers are introduced when semi-active suspensions were first proposed in the early
1970s. In this kind of suspensions the conventional damper is replaced with a controllable damper.
This novel dampers permit altering the damping force based on a control strategy hence, they do
not add energy to the system. Compared with active dampers, semi-active dampers have simple
construction and lower cost [36]. The great manipulation of semi-active dampers appears in
robustness of improving vehicle dynamics (ride comfort and stability).Also, the main advantage of
this damper is the fail-safe operation this means when the semi-active dampers fail due to power
loss, the MR damper still behave as the passive damper [37]. Magnetorheological (MR) fluid
dampers are become trendy in semi-active suspension system as they have the ability to alter the
damping force and to adapt with the environment [13].
Table ‎2.1 Properties of ER vs. MR fluids [10]

Property ER fluids MR fluids
Plastic viscosity 0.2 to 0.3 (Pa.s) 0.2 to 0.3 (Pa.s)
o o
Temperature range +10 to +90 C -40 to +150 C
Power supply (typical) 2 to 5 kV;1 to 10 mA 0.002 to 0.025 kV;1 to 2A
Maximum yield stress 2 to 5 kPa 50 to 100 kPa
Maximum field 4 kV/mm 250kA/m
-7 -8 -10 -11
Plastic viscosity/yield stress 10 to 10 s/Pa 10 to 10 s/Pa
3 3
density 1 to 2 g/cm 3 to 4 g/cm
response time mm sec mm sec
2.6 Magnetorheological fluid dampers

Magnetorheological (MR) dampers are semi-active control devices that exploit MR fluids to create
controllable forces. They can be used as smart actuators for suppressing vibration of suspension
system. MR dampers generally consist of a piston, a cylindrical body filled with MR fluid, an
electromagnet, a pair of electrical wires connected to the electromagnet, a diaphragm, an
accumulator, bearing and seal unit [38]. Figure 2.8, shows a smart and a compact size of MR
damper developed by Lord Corporation which includes a fixed orifice damper filled with MR fluid
37
[33]. In such a damper, fluid is transferred from the high pressure chamber to the low pressure
compartment and vice versa, through orifices in the piston. The accumulator contains pressurized
nitrogen gas, separated from the MR fluid by a diaphragm or a floating piston.
The accumulator has two functions; the first function is to provide more volume for the MR fluid
during the compression stroke when the piston moves into the cylinder. The second function is to
offer pressure equilibrium to prevent cavitations of the MR fluid during the extension stroke when
the piston moves out of the cylinder. The electromagnet produces a magnetic field in the annular
orifice. This magnetic field causes a change in MR properties, which in turn affects the damping
force of the damper.
Figure ‎2.8 Sectional view of MR Damper [22]
There are three major types of MR dampers. These are the mono-tube, the twin tube, and the
double-ended MR damper. The mono-tube damper is shown in Figure 2.9. This damper is
considered as the most common damper used in suspension as it has a compact size and can be
easily mounted in any orientation. The second type of MR damper is the twin tube see Figure 2.10.
Such type has double reservoirs, one contained by the other. The inner cylinder housing contains
the piston, precisely as in a mono-tube damper. It also contains the MR fluid. The outer housing is
moderately filled with MR fluid to compensate for the variations in volume as a result of piston
movement. Regulation of the fluid between the inner and outer reservoir is accomplished by a
valve assembly (usually called foot valve). When the piston inwards into the inner cylinder, the fluid
runs to the outer cylinder during the compression stroke [39]. This process is reversed when the
piston is withdrawn from the damper. The third type is a double-ended damper shown in
Figure 2.11. There are two equal piston rods extending beyond both ends of the damper housing.
There is no accumulator needed in such type of damper, as the piston moves relative to damper
body [39].
Figure ‎2.9 Mono-tube MR Damper [40]
38
Figure ‎2.10 Twin tube MR Damper [40]
Figure ‎2.11 Double-ended MR Damper [40]
2.7 Magnetorheological damper equivalent models

MR dampers have been used to a great extent in vibration control of various dynamic systems
involving vehicle suspension system. For better controlling of unfavourable vibration, a proper
control policy is introduced in conjunction with MR damper. To assess the potential of an MR
damper in different vibration control applications and to exploit the full benefits of the exclusive
features of these devices, a model must be created to accurately imitate the behaviour of an MR
damper [33]. Due to the nonlinear and hysteresis characteristics of MR dampers, a more precise
damper model is needed to achieve perfect damping force alteration and to correctly represent the
inherent hysteresis properties of the damper [41] This will enable the application of various control
strategies to such MR dampers and the fully exploitation of its potential. It is practical to distinguish,
for evaluation of MR hysteresis behaviours, between two types of models; the parametric models
and non-parametric models [10, 42]. Parametric models simulation of MR damper is determined by
adjusting of several parameters which are fitted to experimental results. However, non-parametric
models employ a series of numerically efficient mathematical functions [43].
2.7.1 Parametric models

Parametric models refer to those models which are characterized by numerical relations with
variable coefficients that can be adjusted until the model results directly match the experimental
data. Various parametric models usually comprise springs and dashpots [10]. The most commonly
used model that describes the behaviour of ER and MR fluids is the Bingham model, which
consists of a Coulomb friction element and a dashpot. The configuration of the model contains a
Coulomb element in parallel with the dashpot is shown in Figure 2.12(a). The study of the
behaviour of ER fluid was first carried out by [44] then, a similar model was proposed by [42] to
characterize the MR damping. The robustness of the model depends on the degree of capturing
the non-linear characteristics of the damper. Experimental results show that the model does not
demonstrate the hysteresis behaviour of the damper. This is noticed from the force-velocity
39
response, especially when the acceleration and velocity have opposite signs and the velocities
have small magnitudes [42]. Likewise, a linear relationship of force-velocity is not realistic as
experimental data does not have a linear relationship. Hence, the model cannot be used for
analysis of nonlinear and hysteretic characteristics of MR dampers. The most numerically accurate
and extensively used model to emulate the hysteresis systems is the Bouc-Wen model as depicted
in Figure 2.12 (b). This model is tremendously adaptable and can demonstrate a wide variety of the
hysteresis performance [41-42, 45]. Also, the model is more suitable for numerical simulation as
the dynamic equations are less rigid in comparison to the extended Bingham model [10]. However,
it is observed from [42] that the model does not predict results that are similar to the experimental
results particularly, at the region where the acceleration and velocities have opposite sign and at
small velocity range.
To improve the prediction of the MR hysteresis characteristics and overcome problems in the prior
discussed models, a modified Bouc-Wen model is advocated by [41, 43, 46-52] as shown in
Figure 2.12(c). The recommended model offers a very good prediction of the MR behaviour in the
vicinity of small velocity regions and where the acceleration and velocity have opposite signs [42].
Moreover, the error between the model and experimental data are noticeably smaller than those
computed from other models.
(a) (b)
(c)
(a) Bingham model [53], (b) Bouc-Wen model [42] and (c) Modified Bouc-Wen [42]
Figure ‎2.12 Schematic diagrams for MR damper parametric models
2.7.2 Non-parametric models

When non-parametric models are being developed, no account of any assumptions made.
However, they are entirely derived from the behaviour of a particular rheological fluid device
subjected to different excitations [10]. One of these models, is the polynomial model developed in
[41]. In that model the force-velocity dependence, was divided into two sections which are the
upper and lower corresponding to the negative and positive acceleration respectively. After that the
polynomial function was used to fit the experimental data in these two regions. The proposed
polynomial model predicts fairly well the experimental data with no alteration in the polynomial
40
coefficients [41]. An extra model is suggested in [46, 54] to study the hysteresis behaviour of the
MR damper; this model is called the Non-Parametric Linearized Data Driven model. In this model
the damping force and piston velocity experimental data are divided also into two regimes
analogous to positive and negative accelerations. In each regime, the force and velocity data are
constructed into two linearized vectors in a look up table. The results obtained for this model strictly
matches the experimental hysteresis characteristics of the MR damper [46, 54].
Another nonparametric method, which represents a multi-dimensional curve fitting of the measured
non-linear characteristics of a damper, is the restoring force surface method (RFS) [55]. In this
method, the damping force of a damper is approximated as a function of the displacement and
velocity of the damper using a two dimensional (2D) Chebyshev polynomial [27]. For the
identification of the MR dampers, the damping force is approximated as a function of the
displacement, velocity and voltage as in [52]. Also, the comparison and validation between the
Chebyshev approximation and the measured MR damper characteristics using electrohydraulic test
machine and the simulated characteristics using modified Bouc-Wen model are discussed in [52].
An extra strategy for modelling MR damper is the Neural Network Technology (NNT) that is most
often used due to its ability to model simple as well as very complex function relationships. The
NNT methodology includes first constructing the network architecture, and finally training them. The
advantage of NNT is that it can be used to develop both the direct and inverse model of the MR
damper [49, 51]. In the direct model the output is the damping force and the input is the applied
voltage and the displacement across damper ends. However in the inverse model the output is the
required voltage necessary for producing the desired force [49, 51].
2.8 Semi-Active control of tracked vehicle suspension fitted with MR dampers

In fact, any control system consists of the plant (i.e. the system to be controlled), sensors and a
controller including the hardware and control methods. It is known that a vehicle’s passive
suspension has two conflicting parameters; ride comfort and vehicle handling or stability. These
parameters are preselected by the designer according to the objective and proposed application of
the suspension. Passive suspension is unable to eliminate the trade-off between ride comfort and
stability at all times due to its fixed parameters (i.e. spring and damping characteristics). If the
designer intends to suppress the vehicle’s vibration, the suspension parameters are chosen to be
soft (low damping), however vehicle stability will not be safe. Alternatively, if the suspension is
selected to enhance stability of the vehicle (high damping constant) the user is subjected to large
amounts of vibration and the ride comfort will be poor. Superior design of passive suspension
should eliminate this trade-off between ride comfort and stability or at least try to optimize the
conflicting targets. As a result of the inability of the passive suspension to secure the compromise
between ride comfort and stability, the tendency to use a semi-active suspension is becoming
necessary. In this type of suspension, a controllable damper is used to alter the damping force
consistent with control guidelines which is selected in advance to meet the applications needs of
the suspension.
The semi-active control devices that employ MR dampers cannot input energy into the mechanical
system being controlled [34, 56]. The challenge of controlling these semi-active devices comes
41
from the nonlinear dynamic behaviour of such dampers. It is the command voltage applied to the
current driver for the MR damper prior to the damping force, which can be directly controlled [57].
The robustness of the control method depends on how seriously it can deal with the nonlinear
relationship between the damping force and the applied voltage/current. In tracked vehicle
suspension, semi-active control via MR damper is achieved by two levels of control, namely: the
damper controller and the system controller. The damper controller is exercised to create and
amend the command voltage applied to the damper in order to produce the damping force which is
required to track the desired damping force that is specified by the system controller on the basis of
desired and actual damping forces [57]. The damper and system controller are described in some
details.
2.8.1 Magnetorheological damper controllers

To determine the role of the damper controller, it is reported in [57] that it is used to allow the
damping force to track the desired force demanded by the system controller. It is becoming
increasingly difficult to ignore two factors that have great influence when designing the damper
controller: (1) tracking ability of damping force, and (2) energy requirement for MR damper. The
term tracking ability can be defined as the ability of the controlled damping force to follow the
desired force specified by the system controller. The error between these two forces is used as an
evaluation for this factor. However the term energy requirement is generally understood to denote
the energy required by the system in order to function. Less energy expenditure will extend the
effective life time of the power supply and hence enlarge the life time of the damper. Hereafter the
different types of MR damper controllers are discussed in some details.
The first type of the MR damper controller is the Heaviside step function (HSF). Data from several
sources such as [33-34, 50, 57-58] have identified the HSF as an “on-off” controller where the input
voltage is varied between two values i.e. zero or maximum. In [34] the HSF is utilized to control two
MR dampers in a half car suspension model. The applied voltage here is altered between 0 and
maximum (2 volts) depending on the value of the desired damping force. Still in [58] the same
approach is considered but for controlling one degree of freedom suspension system.
A modified algorithm that alters the command voltage in a discrete way is the signum function (SF)
as described in [47, 57]. This approach varies the input voltage to a certain value between 0 and
maximum. A comparison between the two previous algorithms was made in [57] concluding that
the SF has a better tracking ability and with less energy requirement than the Heaviside method.
Another control method is the continuous state (CS) method which operates in a different manner.
In this method the command voltage is altered continuously with no intermediate value between 0
and maximum. It has been demonstrated that the CS controller accurately tracks the desired force
and with lower energy consumption than the case with the constant voltage of MR damper [18, 58].
Unlike the previous controllers, which are model-based controllers, the non-parametric approaches
such as polynomial function has been applied to illustrate the MR damper’s hysteresis [41, 52]. In
this study a polynomial of sixth order has been used to express the damping force as a function of
piston velocity and the polynomial coefficients are linearized with respect to the input current. The
suggested controller was validated experimentally and compared to Bingham and modified Bouc-
42
Wen models. It has conclusively been shown that the proposed polynomial approach fairly well
predicts non-linear hysteresis behaviour of MR damper under various operating conditions.
However there is a limitation of using the polynomial function as it cannot predict the behaviour of
an MR damper favourably at relatively low velocity. The reason for such deficiency is that the
model lacks the variables that explore the pre-yield aspect of the MR damper [59]. Also the
recurrent neural network (RNN) algorithm was employed to imitate the inverse dynamics of the MR
damper [51, 60]. This is done through voltage estimation to be inputted to the damper so that a
desired damping force can be obtained.
2.8.2 System controllers

In contrast to the damper controller, the system controller is responsible for specifying the desired
damping force according to the dynamic behaviour of the plant [49]. There is a large volume of
published studies describing the role and the development of the system controller for different
control strategies. The applications of the system controller in vibration control isolation, started in
civil engineering applications, it passed on to vehicle suspension control and was extended to
tracked vehicles suspension. There are two basic approaches that are currently being adopted in
research into semi-active control strategies. One is the conventional control strategies and the
other is the modern control strategies. Generally, the conventional and modern control policies
provide many types of control. It is noted that in this work the focus is only on the following types
for conventional control policies:
a- Skyhook control [6, 12, 17, 61-63]

b- Groundhook control [6, 17, 63-64]
c- Hybrid control [6, 17, 63]
However, the modern control policies include:
a- H∞ control method [59, 65-66]

b- Sliding mode control [18, 56, 67-68]
c- Fuzzy logic control [69-76]
There are large number of studies that present a comprehensive simulation analysis of semi-active
suspension using skyhook control scheme and its variant groundhook and hybrid control strategies.
But there exist few studies that offer experimental assessments of such systems. In [63] the prior
three semi-active control policies were experimentally considered in the quarter car suspension
model incorporated with MR damper. In the skyhook algorithm its name comes from a damper
“hooked” between the body (sprung) mass and the “Sky” as revealed in Figure 2.13(a). The
concept of such a control scheme tends to reduce the transmissibility of the sprung mass (i.e. the
ratio between amplitude of the sprung mass and amplitude of the input excitation). In [61] an
extension of exploiting the skyhook control for controlling a full car suspension model featuring four
MR dampers, via hardware-in-the-loop simulation (HILS), was evaluated.
43
(a) Skyhook (b) Groundhook (c) Hybrid
Figure ‎2.13 Idealized skyhook control schemes [63]
The obtained results from the experimental tests show that the intended skyhook control
substantially improves the ride comfort. In contrast to skyhook , a damper configured between the
wheel mass (unsprung) and the “Ground” form the so-called groundhook control, see
Figure 2.13(b). In this case, the damper attempts to diminish the unsprung mass transmissibility.
An alternative semi-active control algorithm identified as hybrid control, merges the concept of both
skyhook and groundhook control to capture the advantages of both control schemes as exposed in
Figure 2.13(c). The hybrid control holds the promise of accomplishing semi-active control policy
that can gradually adapt to various operating conditions and vehicle dynamics desires. The
experimental results obtained from [63] show that the skyhook control affects and reduces the
transmissibility of the sprung mass when compared with passive suspension. This attenuation, for
sure, yields improvement of ride comfort aspect. The results of groundhook control policy are
similar to those of skyhook . However, the transmissibility of the unsprung mass was reduced
resulting in vehicle stability. Nevertheless, all the previously mentioned control algorithms
experience certain drawbacks. None of them can provide transmissibility improvement for both
sprung and unsprung mass simultaneously.
For the hybrid control results, it is expected that these results be similar to the mutual effect of
skyhook and groundhook control and depend on the degree of weighting factors or gains that bias
the performance of hybrid control towards skyhook or groundhook control [63]. Furthermore, a
semi-active static output feedback H∞ control of a quarter car suspension system model with an
MR damper was considered in [59]. The designed scheme employs the measurable suspension
deflection and sprung mass velocity as feedback signals. The intended control scheme was
validated through a numerical simulation under random excitation in the time domain. On average,
there was a significant perfection in the suspension performance [59]. Another work [65] ,adopts a
H∞ controller for a full vehicle suspension model integrated with four MR dampers. The controller
takes into account the system uncertainties and external disturbances. The feasibility of the robust
controller was adopted for evaluation under assorted conditions in both time and frequency domain
during hardware-in-the-loop simulation (HILS). The results of this investigation show that the
suggested control policy is very effective for vibration isolation of vehicles, especially for sprung
mass acceleration and tire deflection.
44
Consistent with [58], the robustness of a sliding mode controller issued from its consideration of
loading uncertainties. Such a controller was applied for one degree of freedom (1DOF) suspension
system with MR damper. After that the controller was applied to control two-degrees-of-freedom (2-
DOF) suspension of a car [18]. In order to evaluate the controller performance, two kinds of
excitation (bump and random) were adopted through a computer simulation in the time and
frequency domains [58]. The most obvious finding to emerge from this study is that the planned
control policy has the ability to reduce both peak and root mean square responses compared to
passive suspension.
Another important control algorithm is the fuzzy logic that permits the use of linguistic variables and
is independent of a fixed mathematical model. This controller was used to control a quarter
suspension model of a mini car [73]. The controller has three fuzzy input variables which are
sprung mass velocity, unsprung mass velocity and the relative velocity across the damper end. The
damping force is the fuzzy output. The results from real tests indicate that the ride comfort and the
driving safety are significantly enhanced [73].
Little studies were found in the literature on semi-active control strategies for tracked vehicles
suspension system. Merely three papers on semi-active control algorithms of tracked vehicle
suspension system featuring ER dampers were found [14, 72, 77]. Another important finding on
intelligent control strategies for semi-active suspension system of tracked vehicles fitted with MR
damper was reported in [14, 19, 70, 74, 78-81]. The application of fuzzy logic control on a 7-DOF
model of tracked vehicle equipped with ER damper and shape memory alloy (SMA) spring was
examined in [14]. The simulation carried out in this research showed that the fuzzy controller
reduces the vibration of the driver’s seat sub ected to random excitation and disturbances. In
another major study [72], mixing between fuzzy logic and sky-groundhook control of tracked vehicle
semi-active suspension with ER suspension unit was studied. The sky-groundhook controller in this
study was applied to 16-DOF models of tracked vehicles taking into account the ride and steering
stability. Adopting of the weighting parameters between the two performance requirements were
achieved by the fuzzy logic controller. The fuzzy input variables were the vertical speed of vehicle
body and rotational angular speed of the wheel and the output variable was the weighting
parameter between the two inputs. Clear benefits of applying fuzzy sky-groundhook controller
could be identified through a computer simulation under bump and random excitation. The
acceleration of the driver’s seat was reduced under bump excitation and ride comfort enhanced
under random disturbances [72].
2.9 Summary
In this chapter, a background of the tracked vehicles suspension types is presented. This includes
both the conventional and non-conventional suspension. A suspension system of a light weight
tracked vehicle M113 is described as an example. The modern suspension systems that include
the active and the semi-active suspension systems were also reported with their advantage and
disadvantages. All types of hydraulic and hydro-gas shock absorbers employed in the existing
tracked vehicles suspension are considered. A full description of the hydraulic and hydro-gas
dampers is presented. A detailed history of smart fluids and magnetorheological dampers are
45
illustrated. This is followed by the different models for the MR dampers obtained from the literature,
including parametric and non-parametric models. Finally, the various control methods for the semi-
active suspension, comprising the MR damper controllers and system controllers are presented.
46
PERFORMANCE UNDER BUMP TERRAINS
3.1 Introduction
In this chapter, the ride dynamic model for the tracked vehicle passive suspension system is
developed based on the differential equations that describe its physical characteristics. Then, the
Simulink toolbox within the MATLAB is used to formulate the block diagrams of the model
equations. In addition, various bump road profiles are used to assess the suspension performance
over different bump road conditions. Furthermore, the developed model is validated using
published experimental field results of APC M113 vehicle. The suspension indices used to evaluate
the suspension performance is presented. A parametric analysis, to demonstrate the influence of
the variations in the vehicle suspension design parameters on the ride dynamic behaviour, is
discussed.
3.2 Half model of passive suspension for tracked vehicle

As stated in the introduction, a variety of mathematical dynamic models for full or half tracked
vehicle suspension models have been developed for the evaluation of the tracked vehicle
performance. However, a two dimensional model of the tracked vehicle is found to be sufficient for
detailed analysis of the suspension dynamics and associated vibration environments [82]. For the
purpose of this study, an in-plane dynamic model of a typical off-road tracked vehicle, APC M113,
is developed in MATLAB/Simulink environment. The model has N+2 degrees-of-freedom (DOF); N
is the number of road wheels per side. Figure 3.1 illustrates a half model of the M113 wherein half
of the hull mass is supported by 5 road wheels. The model is asymmetry (the number of road
wheels in front of the body centre of gravity is greater than the number of wheels behind the body
centre of gravity) about a vertical axis passing through the centre of gravity (C.G) of the hull mass.
In addition, it has 7-DOF, vertical bounce & pitch angle associated with hull mass and 5 DOF
(vertical bounce) of road wheels. The suspension system is modelled as a parallel combination of
linear springs and dampers, wherein the wheels stiffness is represented by equivalent spring
stiffness.
It is noted that the actual suspension of the APC M113 has three hydraulic dampers attached to
wheel numbers 1, 2 and 5. However, there are five dampers shown in the model for the purpose of
investigation. The differential equations governing the suspension model are expressed based on
Newton’s Second Law of Motion. Also, in the development of the model, the following assumptions
are made:
1- The hull body mass element is assumed to be rigid body.

2- The torsion bars are represented by independent linear springs.
3- The shock absorbers represent the damping elements which dissipate the energy and have
constant damping coefficients.
4- The road wheel tyres are assumed to be much more rigid than the torsion bars;
47
5- The effect of the track on the dynamics of the suspension is not considered.
Figure 3.1 Half tracked vehicle suspension model
Based on these assumptions, the half suspension system of the tracked vehicle model can be
developed. The hull bounce, pitch and the wheel bounce motions are expressed in the following
equations and the data provided for simulation are taken from [82-84] and listed in Table 3.1.
Bounce motion of the hull:
θ θ (‎3.1)
Pitch motion of the hull:
θ θ θ (3.2)
th
Bounce of i road wheel:
- θ- - θ- - ; for i=1,..., 5 (3.3)
All the variables and parameters in these equations are defined in Table 3.1 which also states their
typical values used in the simulations.
48
Table ‎3.1 Half model suspension parameters for tracked vehicle [82-84]
Description Symbol Values Description Symbol values
st a
Body mass (kg) 5109 1 wheel centre (m) 1.35
nd
Body moment of inertia 12856 2 wheel centre (m) 0.69
2
about lateral axis 'y' (kg.m )
th rd
Equivalent stiffness of i 104000 3 wheel centre (m) 0.02
torsion bar (N/m)
th th b
Damping coefficient of i 22520 4 wheel centre (m) -0.66
damper (N.s/m)
th th
Mass of i road wheel (kg) 113.5 5 wheel centre (m) -1.32
th
Equivalent stiffness of i 613000
wheel (N/m)
a: The horizontal distance between the hull centre of gravity and the road wheel centre
b:Negative values denote that the distance is measured behind of the body centre
3.3 Bump excitations

The bump road profiles are characterized for the evaluation of the transient response of most of the
tracked vehicles as in reference [85]. Three different road profiles namely as; the shallow, medium
and sharp bump terrains are considered in this study. The road profiles are selected to cover a
wide range of different operating conditions. The vertical excitation of the first road wheel is
characterized by the following equation:
(3.4)
; (3.5)
where h represents the bump height; w represents the bump width; r represents the angular
frequency and represents the vehicle speeds. In addition, the vertical excitations to the other
wheels r2 r are described by:
(3.6)
where the time delay between the bump excitation of wheel 1 and the subsequent bump excitation
of wheel is given by:
, i=2,..., 5 (3.7)
Figure ‎3.2 shows the selected bump road profiles which are applied at the first road wheel. The
bump has a height of 0.1 m and three different bump widths according to the type of the road
profile. Table ‎3.2 lists the parameters of the bump road profile characteristics.
Table ‎3.2 Description and parameters of selected road profiles

Road profile type Height (m) Width (m)
Shallow 5
Medium 0.1 2.5
Sharp 0.5
49
0.1
Shallow
0.08 Medium
Amplitude (m)
Sharp
0.06
0.04
0.02
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Time (s)
Figure ‎3.2 Typical bump road profiles at vehicle speed of 10 km/h
3.4 Simulink model

This section presents the Simulink model of the passive suspension system of the tracked vehicle.
The model is developed directly from the differential equations (3.1, 3.2 and 3.3) that describe the
physical model. The Simulink toolbox within MATLAB is used to build the block diagrams that
describe the model equations. The suspension model shown in Figure 3.3 represents the top-level
diagram of the passive suspension system which includes the suspension and damper forces and
moments. The top-level diagram is used to model Equations (3.1) and (3.2). The equations are
implemented in the Simulink through the straightforward use of Gain and Summation blocks to give
the body bounce and angular motions.
There are five suspension stations marked by cyan on the model diagram representing Equation
(3.3). These suspension stations are masked and their under mask block diagram is shown in
Figure 3.4. The input shown in green on the model diagram is applied to each road wheel in
addition to the suspension forces to give the vertical wheel motions.
50
Figure 3.3 Top-level diagram of the passive suspension model
51
Figure ‎3.4 Under mask block diagram for first suspension station
52
3.5 Computer model validation
The dynamic response of the undamped tracked vehicle suspension is evaluated using the
MATLAB program in order to calculate the associated natural frequencies. The free-undamped
system principle is used to calculate the model natural frequencies using the following equation:
(‎3.8)
The above equation contains the mass and stiffness matrices (M and K) as well as the response
vector z(t). These matrices are illustrated below. In order to obtain the natural frequencies, the free
vibration (damping and excitation forces are neglected) is considered in which the determinant of
should equal zero. Table 3.3 lists the calculated natural frequencies.
- - - - -
2
i li kbi i li kbi - - - - -
- - k
- - k2
- - k3
- - k4
- - k
, , , ,
where: , , , ,
and the displacement and acceleration vector are expressed as follows:
T
and b w w2 w3 w4 w
Then, the predicted natural frequencies of the system are directly compared against a field
measurement data of a real tracked vehicle. The predictions are carried out for two sets of hull
masses denoted as laden and unladen vehicle. Table 3.4 lists the predicted and measured natural
frequencies of the system response and the corresponding modes of vibration. The measured
frequencies are also for laden and unladen M113 tracked vehicles [86]. The results show good
agreement between the predicted and measured frequencies.
53
Table ‎3.3 Predicted natural frequencies of 7-DOF suspension for laden vehicle
Dominant degree of freedom type Undamped Natural frequencies (Hz)

Hull bounce 1.48
Hull pitch 0.88
st
1 wheel bounce 12.64
nd
rd
th
th
Table ‎3.4 Predicted and measured undamped natural frequencies and associated modes
Undamped natural frequency (Hz)
Mode
Current work Reference [86] Mode type
No.
(laden) (unladen) (laden) (unladen)
1 0.88 1.24 1.23 1.31 Hull pitch
2 1.48 1.75 1.65 1.85 Hull bounce
3 12.64 11.64 11.39 11.32 Wheel #1 bounce
4 12.64 12.64 12.2 11.08 Wheel #2 bounce
5 12.64 12.64 12.26 12.05 Wheel #3 bounce
6 12.65 12.66 12.34 12.01 Wheel #4 bounce
7 12.66 12.67 11.97 11.41 Wheel #5 bounce
Laden vehicle with hull mass 5109 kg; unladen vehicle with hull mass 3660 kg
3.6 Simulation procedures

Three vehicle suspension configurations namely: configuration A, B and C, are used in the
simulation in order to evaluate the vehicle suspension performance under various conditions. In
configuration A, the suspension has no dampers i.e. undamped suspension and configuration B
contains different number and locations of the damper. Table 3.5 summarizes the possible
suspension configurations for vehicle configuration A and B. The table includes five suspension
configurations in addition to the undamped suspension; these configurations are denoted as
configuration A (S0) and configuration B ( S1, S2, S3, S4 and S5). Configuration S0 includes no
dampers, S1 contains one damper with five alternative locations, and S2 includes two dampers
with ten alternative locations. Also, suspension configuration S3 has three dampers with ten
locations, S4 has four dampers with five different locations and S5 has five dampers. All these
configurations are selected primarily to provide a sufficiently wide range for the investigation of the
effect of the suspension settings on the tracked vehicle performance. Not all the configurations may
have applications in practice. Finally, vehicle configuration C is the standard vehicle suspension
configuration which includes three dampers fitted at the first, second and fifth road wheel. There
are several suspension indices that can be employed while evaluating the suspension performance
of tracked vehicles. Bounce acceleration (BA) and angular acceleration (AA) are the most used
indices [19, 25, 72, 77]. In addition, however, the hull vertical displacement and pitch are also used
to assess suspension performance. Furthermore, the most critical values in a suspension design
process are the maximum amplitude which is related to the crest factor (CF) that must also be
considered in the design [87]. Therefore, in order to assess and compare the performance of
54
different configurations of a tracked vehicle suspension system, it is necessary to introduce a
metric that combines the vertical and pitch acceleration crest factors and the vertical and angular
displacement crest factors into a single variable. In this thesis, the variable used is referred to as
the Mixed Objective Function (MOF). It should be noted that this is the first time that such a
variable has been used. Similarly, this is the first time that a systematic study has been conducted
on tracked vehicle suspension optimisation based on the number and location of dampers. Hence,
in chapters 3 and 4, assessment of the suspension performance was carried out by using the MOF
that combines both the hull vertical bounce and pitch angle. This combination uses the Crest
Factor (CF) values of bounce acceleration (BA), bounce displacement (BD), angular acceleration
(AA) and pitch angle (PA) to give an indication about the ride comfort improvement. The MOF
enables the assessment of the vehicle ride performance for the thirty alternative suspension
configurations shown in Table 3.5. These configurations include various numbers of dampers at
different damper locations in the suspension. The equation that describes the MOF function can be
expressed as follows:
4 2
MOF i Fi ; F Ai,peak Ai, ; Ai BA,B ,AA, and PA (3.9)
Table ‎3.5 Proposed various suspension configurations
Suspension No. of
dampers Damper locations in the suspension
confg.
S0 - - - - - -
S1 1 wh1 wh2 wh3 wh4 wh5
wh12 wh13 wh14 wh15 -
wh23 wh24 wh25 - -
S2 2
wh34 wh35 - - -
wh4,5 - - - -
wh123 wh124 wh125 - -
wh134 wh135 - - -
wh145 - -
S3 3
wh234 wh235 - -
wh245 -
wh345 - - - -
wh1234 wh1235 - -
wh1245 - - - -
S4 4
wh1345 - - - -
wh2345 - - - -
S5 5 All wheels
55
3.7 Parametric analysis of suspension performance under bump terrains
In general, the vehicle dynamic behaviour is influenced by a number of factors, such as vehicle
speed, road profile and suspension system. A parametric study is helpful to understand the effect
of these different parameters on the vehicle dynamic performance. In addition, it provides a
foundation for selecting the optimal parameters that offer the best ride. The objective of the
parametric study is to demonstrate the influence of variations in the passive suspension
parameters on the ride dynamic behaviour of the tracked vehicle. The parameters considered for
the variations are: locations and numbers of dampers, equivalent damping coefficient, equivalent
suspension stiffness and road wheel stiffness. The vehicle configurations A and B are considered
for the locations and numbers of dampers analysis and the vehicle configuration C is used for the
other analysis. All the analyses have performed under the three different road profiles which are
discussed earlier. It should be noted that tracked vehicle suspension systems are different from
road vehicle suspension systems. The presence of multiple road wheels in the suspension system
affects the vehicle dynamic performance. Hence, understanding the effect of road excitation types
and vehicle speeds on the performance of tracked vehicles is important. The following section 3.7.1
presents the performance of a tracked vehicle traversing various bump excitations with speeds of
10 to 60 km/h. Sections 3.7.3 to 3.7.6 presents the influence of the number, locations of hydraulic
shock absorber, damping coefficient, suspension and wheel stiffnesses on the performance of the
tracked vehicle.
3.7.1 Effect of speed on dynamic response of tracked vehicles traversing a bump

In this section, the performance of the standard tracked vehicles (five road wheels and three
st nd th
hydraulic dampers located at 1 , 2 and 5 wheel stations) traversing shallow and sharp bump
road profiles is studied. The vehicle crosses the road profile with constant speeds of 10 to 60 km/h.
The bounce acceleration (BA) and angular acceleration (AA) are used as performance criteria to
assess the tracked vehicle dynamic response.
Figure 3.5 and Figure 3.6 show the time and frequency domains of the BA and AA responses
under shallow bump excitation. In these figures, the first column shows the time domain responses
while the second column shows the frequency domain responses. Also, the graphs on the same
horizontal level are for the same speed which varies from 10 km/h to 60 km/h in steps of 10 km/h.
The figures show that the vehicle speed affects the performance of the tracked vehicle suspension.
The time domain plots show that the peak amplitudes of the BA and AA responses are increased
when the vehicle speed is increased. Moreover, the frequency domain plots, which are the FFT
responses, show that maximum peaks are influenced by both the excitation due to the bump profile
as well as the vehicle speed. For example, Figure 3.5 shows that for the shallow bump, the first
peak of each of the FFT plots of the bounce acceleration (BA) at different speeds is dominant, that
is it has the highest value. The magnitude of this dominant peak and the frequency at which it
occurs increase as the speed increases. In addition, extra (secondary) peaks of decreasing
magnitude appear in the FFT plots of BA after the dominant peak. At first, this may appear to be
side lobes associated with leakage phenomenon in the FFT procedures. However, repetition of the
FFT procedures using smaller frequency steps gave exactly the same results. A careful
56
examination reveals that the secondary peaks are due to the dynamics of the multiple road wheels
crossing the bump terrain.
For example, Figure 3.5 shows that at a speed of 10 km/h, there is a dominant peak in the FFT plot
followed by about nine peaks of reducing peak amplitudes. As the speed is increased, the number
of secondary peaks decreases. The figure shows that the total number of peaks in the FFT plots
decreases progressively from 10 peaks at 10 km/h to 3 peaks at 60 km/h. On the other hand,
Figure 3.6 shows that for the shallow bump, the FFT plots of the angular acceleration (AA) have
one dominant peak whose magnitude decreases generally with speed while its frequency band
increases with speed. Also, the frequency of the peak increases with speed but is affected by the
pitch and bounce natural frequencies of the vehicle which occur at 0.9 and 1.5 Hz, respectively.
Similarly, the BA and AA responses of the vehicle traversing a sharp bump profile at the same
speeds are shown in Figure 3.7 and Figure 3.8. In fact, the sharp road profile imposes severe
excitation to the suspension system even at low vehicle speeds. This can be seen in the time
domain responses where the amplitude of the BA and AA are significantly increased compared
with the amplitude due to the shallow bump profile. The effect of the vehicle speed on the
performance of the vehicle suspension is also shown in the figures. Moreover, the FFT responses
show multiple peaks at low speeds (10 and 20 km/h) while at high speeds only the wheel
resonance frequency of about 12.5 Hz is clearly shown. The FFT plots also show that the dominant
peaks occur at a higher frequency when the vehicle traverses the sharp bump in comparing to the
shallow bump.
It can be concluded that the dynamic performance of tracked vehicles is influenced by different
parameters such as the vehicle speed and the type of the road profiles. Hence, in order to increase
the vehicle limiting speed, the suspension performance over rough terrains should be improved.
57
0.3 0.06
Speed 10 km/h
0.2 0.05
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.1
0.04
0
0.03
-0.1
0.02
-0.2
-0.3 0.01
-0.4 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Time (s) Frequency (Hz)
2 0.14
Speed 20 km/h
1.5 0.12
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.1
1
0.08
0.5
0.06
0
0.04
-0.5 0.02
-1 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
3 0.2
Speed 30 km/h
2
0.15
Bounce Acc (m/s2)
Bounce Acc (m/s2)
1
0.1
0
0.05
-1
-2 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
4 0.25
Speed 40 km/h
3
0.2
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
1 0.15
0
-1 0.1
-2
0.05
-3
-4 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.35
4 Speed 50 km/h
0.3
3
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2 0.25
1 0.2
0
-1 0.15
-2 0.1
-3
0.05
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
6 0.35
Speed 60 km/h
4 0.3
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2 0.25
0 0.2
-2 0.15
-4 0.1
-6 0.05
-8 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Figure ‎3.5 BA responses of suspension under shallow bump excitations at different speeds
58
0.8 0.14
Speed 10 km/h
0.6 0.12
Angular Acc (rad/s2)

0.4 0.1
0.2 0.08
0 0.06
-0.2 0.04
-0.4 0.02
-0.6 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
3 Time (s) 0.35 Frequency (Hz)
Speed 20 km/h
0.3
2

0.25
1 0.2
0 0.15
0.1
-1
0.05
-2 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
4 0.35
Speed 30 km/h
3 0.3
2

0.25
1
0.2
0
0.15
-1
0.1
-2
-3 0.05
-4 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
4 0.25
Speed 40 km/h
2 0.2
0 0.15
-2 0.1
-4 0.05
-6 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.25
4 Speed 50 km/h
3 0.2
2
1 0.15
0
-1 0.1
-2
-3 0.05
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Speed 60 km/h
2
0.2
0
0.15
-2
0.1
-4
0.05
-6
-8 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Figure ‎3.6 AA responses of suspension under shallow bump excitations at different speeds
59
8 0.5
Speed 10 km/h
6
0.4
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2 0.3
0
-2 0.2
-4
0.1
-6
-8 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
10 0.7
Speed 20 km/h
0.6
5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.5
0.4
0
0.3
0.2
-5
0.1
-10 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
15 5
Speed 30 km/h
10 4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

5 3
0 2
-5 1
-10 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
10 3.5
Speed 40 km/h
3
5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2.5
2
0
1.5
1
-5
0.5
-10 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
10 1.4
Speed 50 km/h
8 1.2
6
Bounce Acc (m/s2)
Bounce Acc (m/s2)
1
4
0.8
2
0.6
0
0.4
-2
-4 0.2
-6 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
10 0.2
Speed 60 km/h
8
6 0.15
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2 0.1
-2 0.05
-4
-6 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Figure ‎3.7 BA responses of suspension under sharp bump excitations at different speeds
60
4 0.16
Speed 10 km/h
3 0.14
2 0.12

1 0.1
0 0.08
-1 0.06
-2 0.04
-3 0.02
-4 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.7
4 Speed 20 km/h
0.6
3

2 0.5
1 0.4
0
-1 0.3
-2 0.2
-3
0.1
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.7
4 Speed 30 km/h
0.6
3

2 0.5
1 0.4
0
-1 0.3
-2 0.2
-3
0.1
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
6 0.7
Speed 40 km/h
0.6
4
0.5
2 0.4
0 0.3
0.2
-2
0.1
-4 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.7
4 Speed 50 km/h
0.6
3
2 0.5
1 0.4
0
-1 0.3
-2 0.2
-3
0.1
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
5 0.5
4 Speed 60 km/h
3 0.4
2
1 0.3
0
-1 0.2
-2
-3 0.1
-4
-5 0
0 2 4 6 8 10 0 2 4 6 8 10 12 14
Figure 3.8 AA responses of suspension under sharp bump excitations at different speeds
61
3.7.2 Influence of damper locations
In this part, the effect of varying the damper locations on the vehicle suspension performance is
investigated. The suspension performance is evaluated under different bump road excitations and
vehicle speeds. It is noted that in this analysis, the damping level is kept constant at a typical
damping constant of M113-A3 tracked vehicle see Table 3.1. The MOF values of all suspension
configurations listed in Table 3.5 are calculated under shallow, medium and sharp bump terrains
and the data are listed in Tables 3.6 to 3.9.
Table 3.6 shows the MOF values of suspension configurations S1 under these excitations. The
results show that the damper locations have a significant effect on the vehicle performance. Also,
the suspension configurations S1 have lower MOF values than the undamped suspension for the
three bump road profiles. Over the shallow bump and vehicle speed of 10 and 40 km/h, the
suspension configuration S1 has two locations with the lowest MOF values which are at wheel 1
and wheel 5 respectively. In addition, when the vehicle traverses the same road with 60 km/h, the
suspension fitted with a damper at wheel 1 or wheel 2 offers the lowest MOF values. Over the
medium bump profile and vehicle speed of 10 km/h, the lowest MOF values occur when the
suspension is equipped with dampers at wheel 1 and wheel 5. For the same road, when the vehicle
speed is increased to 40 and 60 km/h, the suspension with dampers at wheel 1 or wheel 2 has the
lowest MOF values. Over the sharp bump profile with speeds 10 and 40 km/h, the best locations
are at wheel 1 and wheel 5 while at speed 60 km/h, the best damper locations are at wheel 1 and
wheel 3.
Table 3.7 shows the MOF values of the suspension configurations S2 under the same excitations.
The results indicate the significant effect of the damper locations on the vehicle performance. Over
the shallow bump and a speed of 10 km/h, the suspensions with dampers at wheels 13 and wheels
12 are the best configurations that have the lowest MOF values. From this point onwards, it should
be noted that wheels xyz denote wheel locations number x, y and z. Thus, wheels 13 and wheels
12 mean wheels 1, 3 and wheels 1, 2 respectively. When the vehicle traverses the same road with
speed 40 km/h, the best suspension configurations are at wheels 14 and wheels 15. At 60 km/h,
dampers at wheels 15 and wheels 12 have the lowest MOF values. Over the medium bump, the
suspension with dampers at wheels 14 and wheels 15 has the lowest MOF values at 10 km/h while
dampers at wheels 13 and wheels 14 have the best locations at speed 40 km/h. At 60 km/h,
dampers at wheels 12 and wheels 13 have the best locations that have the lowest MOF values.
Over the sharp bump and at speeds of 10 and 40 km/h, dampers at wheels 14 and wheels 13 are
the best locations, while dampers at wheels 15 and wheels 12 have the best damper locations at
speed 60 km/h.
Likewise, the MOF of all suspension configurations S3 are shown in Table 3.8. Over the shallow
bump, the damped suspension has a better performance than the suspension S0. At 10 km/h, the
suspensions with dampers at wheels 245 and wheels 123 have the lowest MOF while at 40 km/h,
the suspensions with dampers at wheels 145 and wheels 135 have the best locations that offer the
better performance.
62
Table ‎3.6 MOF values of suspension configurations S1 under different bump excitations
Suspension Shallow Medium Sharp
confg. V1 V2 V3 V1 V2 V3 V1 V2 V3
undamped 0.196 0.33 0.297 0.243 0.314 0.197 0.261 0.294 0.113
Wh1 0.142 0.106 0.096 0.138 0.081 0.065 0.085 0.063 0.036
Wh2 0.166 0.122 0.106 0.166 0.094 0.074 0.127 0.13 0.103
Wh3 0.198 0.129 0.127 0.163 0.127 0.1 0.124 0.143 0.040
Wh4 0.171 0.118 0.116 0.172 0.119 0.101 0.102 0.132 0.123
Wh5 0.154 0.111 0.110 0.158 0.116 0.096 0.088 0.089 0.054
Table ‎3.7 MOF values of suspension configurations S2 under different bump excitations

undamped 0.196 0.33 0.297 0.243 0.314 0.197 0.261 0.294 0.113
Wh12 0.147 0.111 0.087 0.154 0.085 0.066 0.089 0.092 0.06
Wh13 0.145 0.107 0.090 0.151 0.077 0.066 0.082 0.091 0.076
Wh14 0.148 0.099 0.091 0.128 0.085 0.069 0.082 0.090 0.079
Wh15 0.155 0.091 0.086 0.129 0.087 0.073 0.088 0.096 0.046
Wh23 0.170 0.116 0.095 0.169 0.092 0.072 0.100 0.113 0.082
Wh24 0.151 0.108 0.094 0.164 0.090 0.073 0.097 0.096 0.072
Wh25 0.148 0.100 0.093 0.14 0.096 0.076 0.086 0.109 0.094
Wh34 0.171 0.111 0.106 0.151 0.108 0.087 0.095 0.118 0.094
Wh35 0.156 0.104 0.104 0.135 0.108 0.085 0.085 0.100 0.091
Wh45 0.156 0.106 0.105 0.151 0.109 0.084 0.089 0.111 0.086
V1= 10 km/h; V2= 40 km/h; V3= 60 km/h
For the medium bump and at 10 km/h, the best locations for the dampers are at wheels 124 and
wheels 235, while at medium speed of 40 km/h, the best locations are at wheels 124 and wheels
125. At 60 km/h, the best damper locations are at wheels 124 and wheels 123. For the sharp bump
at 10 km/h, wheels 134 and wheels 235 have the best damper locations, while the best locations at
40 km/h are at wheels 124 and wheels 135. At 60 km/h, the best damper locations are at wheels
123 and wheels 145.
Table 3.9 lists the MOF values of the suspension configurations S4 and S5. Also, all the damped
suspension configurations have lower MOF values than the undamped suspension S0. Over the
shallow bump and at 10 km/h, the suspension S4 has two locations with the lowest MOF which are
at wheels 1245 and wheels 2345 while dampers at wheels 1245 and wheels 1345 have the best
locations at speed 40 km/h. At high speed, dampers at wheels 1245 and 1345 have the lowest
MOF values. Over the medium bump, wheels 1345 and wheels 1234 are the best damper locations
at 10 km/h while wheels 1235 and wheels 1345 are the best locations at 40 km/h. At speed of 60
km/h, wheels 1234 and wheels 1235 are the best locations.
It can be seen from the results that the damper locations affect the vehicle performance and the
best damper locations over specific road profile are changed when the vehicle speed is varied. In
order to select the best damper locations that suit different bump road profiles, the MOF values for
all speeds are summed together. Then, the summed MOF values for each road are added together
to give a resultant MOF per each suspension configuration. After that, the suspension with the
63
lowest resultant MOF value is considered the best damper location that offers good suspension
performance as presented in Table 3.10. The results show that the damper locations have a
significant effect on the vehicle performance. The resultant MOF gives an indication of the best
locations. From the results, there are five suspension configurations with the lowest MOF values
which are S1-1, S2-15, S3-125, S4-1245 and S5-all. It is noted that suspension with dampers at
wheels 125 are chosen instead of damper locations at wheels 145 because they have relatively the
same MOF. In addition, the suspension configuration with dampers at wheel 125 is a standard
suspension setting of typical off-road vehicles.
Table 3.8 MOF values of suspension configurations S3 under different bump excitations
undamped 0.196 0.33 0.297 0.243 0.314 0.197 0.261 0.294 0.113
Wh123 0.148 0.107 0.086 0.147 0.082 0.066 0.088 0.092 0.042
Wh124 0.149 0.096 0.085 0.127 0.080 0.066 0.082 0.058 0.078
Wh125 0.152 0.091 0.080 0.131 0.080 0.07 0.088 0.077 0.047
Wh134 0.149 0.097 0.086 0.139 0.083 0.069 0.077 0.090 0.052
Wh135 0.159 0.090 0.082 0.130 0.081 0.072 0.092 0.067 0.048
Wh145 0.154 0.087 0.080 0.128 0.082 0.070 0.083 0.073 0.045
Wh234 0.156 0.104 0.089 0.136 0.089 0.072 0.089 0.096 0.066
Wh235 0.151 0.099 0.088 0.128 0.088 0.073 0.081 0.067 0.048
Wh245 0.146 0.099 0.089 0.141 0.090 0.076 0.081 0.069 0.086
Wh345 0.156 0.101 0.097 0.144 0.099 0.081 0.087 0.111 0.054
V1= 10 km/h; V2= 40 km/h; V3= 60 km/h
Table 3.9 MOF values of suspension configurations S4 and S5 under different bump excitations

undamped 0.196 0.33 0.297 0.243 0.314 0.197 0.261 0.294 0.113
Wh1234 0.152 0.098 0.083 0.131 0.079 0.067 0.089 0.086 0.069
Wh1235 0.157 0.091 0.079 0.133 0.077 0.067 0.089 0.072 0.060
Wh1245 0.151 0.087 0.077 0.135 0.083 0.068 0.090 0.057 0.046
Wh1345 0.157 0.088 0.078 0.129 0.077 0.069 0.088 0.075 0.059
Wh2345 0.152 0.099 0.083 0.133 0.086 0.074 0.083 0.094 0.078
All wheels 0.157 0.089 0.076 0.140 0.078 0.068 0.096 0.049 0.039
V1= 10 km/h; V2= 40 km/h; V3= 60 km/h
Table 3.10 MOF for various suspension configurations under different bump excitations
Damper Sum of MOF for all speeds Sum of MOF
Suspension confg.
locations Shallow Medium Sharp for all roads
S0 Undamped 0.669 0.755 0.824 2.248
At wh1 0.186 0.286 0.345 0.817
S1
At wh5 0.232 0.371 0.377 0.980
At wh14 0.253 0.283 0.339 0.875
S2
At wh15 0.231 0.290 0.333 0.855
At wh125 0.213 0.282 0.325 0.820
S3
At wh145 0.202 0.281 0.322 0.806
At wh1245 0.194 0.288 0.316 0.799
S4
At wh1345 0.223 0.277 0.324 0.824
S5 At all wheels 0.186 0.287 0.324 0.797
64
3.7.3 Influence of number of dampers
This section presents the effect of the number of dampers in the suspension on the vehicle
performance. The time and frequency domains of the bounce acceleration (BA) and angular
acceleration (AA) responses of the best suspension locations that have been selected in
section 3.7.2 are presented and compared with the undamped response under the same
excitations and speeds.
1- Shallow bump responses
The responses of the suspension system with the best configurations under the shallow bump
excitation are shown in Figure 3.9(a, b). The time and frequency domains responses of BA and AA
responses of all suspension configurations are compared with those for the undamped suspension
at different speeds of 10, 40 and 60 km/h. The results from Figure 3.9(a) show that the damped
suspension dissipates the energy caused by the shallow bump excitation, reduces the settling time
and improves the suspension performance at all speeds.
Figure 3.9(b) shows the absolute values of the FFT responses of the BA responses under the
same excitations. At the low speed of 10 km/h, the undamped response clearly shows the pitch
and bounce peak amplitudes at frequencies of 0.9 Hz and 1.5 Hz, respectively. For the damped
suspensions, the peak amplitude of the pitch response is practically the same as the undamped
suspension response but the bounce peak amplitudes are considerable suppressed by the damped
suspension. The other peaks seen in the FFT plot at the speed of 10 km/h are due to the dynamics
of the tracked vehicle as described in section 3.7.1. At the higher speeds of 40 km/h and 60 km/h,
the pitch response is not excited because the excitation frequencies are 2.22 Hz and 3.33 Hz,
respectively. In the undamped suspension case, this induces the bounce response of the tracked
vehicle at 1.5 Hz. In the damped suspension cases, this induces a bounce response at frequencies
between the bounce of 1.5 Hz and the corresponding excitation frequency. Also, the damped
suspensions have significantly reduced the BA resonant peak at 1.5 Hz under all speeds.
The PTP values of the BA of all suspension configurations are listed in Table 3.11. The data
indicate that the amount of the damping force and the vehicle speed have a significant effect on the
suspension performance. For instance, at 10 km/h the suspension configuration S4-1245 and S5-
all have the highest reduction in the PTP values, followed by S3-125, S2-15 and S1-1. When the
vehicle negotiates the same road at 40 and 60 km/h, the road-induced vibration transmitted to the
vehicle body is increased. Therefore, the suspension should isolate the vehicle body from these
vibrations. One would have expected that increasing the number of dampers in the suspension will
suppress the road vibration. But, the results from the table confirm that at 40 and 60 km/h,
increasing the damping degrades the PTP responses and increasing the number of dampers over
three reduces the vehicle performance. The RMS values of the BA responses are also listed in
Table 3.11. The results from the table show that the configurations S4-1245 and S5-all have the
lowest RMS values under the shallow bump with 10 km/h. However, over the medium and sharp
bumps, the suspension configurations S1-1 and S3-125 are superior in enhancing the vehicle
performance.
65
Similarly, Figure 3.10(a, b) illustrates the time and frequency domains of the AA responses under
the same excitations. The results shown in Figure 3.10(a) demonstrate that the damped
suspension configurations have a better performance than the undamped suspension S0 for all
speeds. Figure 3.10(b) show that the undamped AA response has a peak amplitude at 0.9 Hz at all
speeds and that the damped suspensions have reduced the AA resonant peaks at 0.9 Hz at all
speeds. Also, the frequencies of the peaks of the damped response are speed dependant. At 10
km/h, the peaks lie between the excitation frequency of 0.56 Hz and the pitch frequency of 0.9 Hz.
But at 40 km/h and 60 km/h, the peaks occur at frequencies between the pitch frequency of 0.9 Hz
and the excitation frequencies of 2.22 Hz and 3.33 Hz, respectively.
Table 3.12 shows the PTP values of the AA responses under the same excitations and speeds.
The table indicates that the damped suspensions have better responses than the undamped
suspension. The suspension configurations S5-all and S4-1245 have the biggest reductions in the
PTP values, followed by the configurations S3-125, S2-15 and S1-1. However, the PTP values of
the AA responses at 40 and 60 km/h are increased when the number of dampers is increased. This
confirms that increasing the damping force is not always beneficial for vibration isolation. The table
also shows that all the suspension configurations have reduced the RMS values of the AA
responses. However, suspension configurations S1-1 and S2-15 have the highest reduction in
RMS values of AA at 40 and 60 km/h. It is to be noted that over the shallow bump road, the number
of dampers have a significant effect on the suspension performance. For example, at low speed,
when the number of dampers is increased, the BA response is improved. However, at medium and
high speeds, the BA response becomes worse as the number of dampers is increased. Also, at
medium and high speeds, the suspensions S1-1, S2-15 and S3-125 offer the optimal settings that
provide the best reduction in the BA peaks over shallow bump. However, the AA responses are
increased when the number of dampers is increased.
66
Speed 10 km/h
0.4 0.16
Undamped Undamped
At wh1 At wh1
0.3 At wh15 0.14 At wh15
At wh125 At wh125
At wh1245 At wh1245
0.2 At all wheels 0.12 At all wheels
Bounce Acc (m/s2) 0.1 0.1
Bounce Acc (m/s2)

0 0.08
-0.1 0.06
-0.2 0.04
-0.3 0.02
-0.4 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
10 9
Undamped Undamped
At wh1 At wh1
8 8
At wh15 At wh15
At wh125 At wh125
6 At wh1245 At wh1245
7
At all wheels At all wheels
4
6
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2
5
0
4
-2
3
-4
2
-6
-8 1
-10 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
Speed 60 km/h
10 9
Undamped Undamped
At wh1 At wh1
At wh15 8 At wh15
At wh125 At wh125
5 At wh1245 At wh1245
7
6
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0
5
4
-5
2
-10
-15 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.9 Bounce acceleration responses under shallow bump excitation
67
Speed 10 km/h
1.5 1
Undamped Undamped
At wh1 At wh1
0.9
At wh15 At wh15
1 At wh125 At wh125
At wh1245 0.8 At wh1245
0.7
0.5

0.6
0 0.5
0.4
-0.5
0.3
0.2
-1
0.1
-1.5 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 1.4
Undamped Undamped
At wh1 At wh1
3 At wh15 At wh15
1.2
At wh125 At wh125
At wh1245 At wh1245
2
1
1
0.8
0
-1 0.6
-2
0.4
-3
0.2
-4
-5 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.7
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
4 0.6
At wh125 At wh125
At wh1245 At wh1245
2 0.5
0 0.4
-2 0.3
-4 0.2
-6 0.1
-8 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
Figure 3.10 Angular acceleration responses under shallow bump excitation
68
2- Medium bump responses
Over the medium bump road, the excitation frequency which is defined as the ratio of the vehicle
speed to the bump width is increased compared to the shallow bump road. The increase in the
road frequency will introduce high vibration levels to the suspension system. Figure 3.11(a, b)
shows the system responses of the BA of all suspension configurations. From Figure 3.11(a), the
results indicate that all the damped suspension configurations have better acceleration responses
than the undamped suspension. Figure 3.11(b) shows the frequency domain responses of all
suspension configurations under the same excitations. At speed of 10 km/h, the excitation
frequency is 1.11 Hz. In the undamped response case, this induces the bounce peak amplitude at
a frequency of 1.5 Hz. The pitch response is not excited. For the damped suspensions, the bounce
peak amplitudes are considerably suppressed by the damped suspension. The other peaks seen in
the FFT graph at the speed of 10 km/h are due to the dynamics of the tracked vehicle as described
in section 3.7.1. At the speeds of 40 km/h and 60 km/h, the pitch response is not excited because
the excitation frequencies are 4.44 Hz and 6.66 Hz, respectively. In the undamped suspension
case, this induces the bounce response of the tracked vehicle at 1.11 Hz. In the damped
suspension cases, this induces a bounce response at frequencies between the bounce of 1.11 Hz
and the corresponding excitation frequencies of 4.44 Hz and 6.66 Hz. It can be seen that the
damped suspension reduces the BA resonant peaks at the resonant frequency of 1.5 Hz, while the
suspension S0 has a better performance than the damped suspension over the frequency range
from 2 to 6 Hz.
The PTP values of the BA for all suspension configurations at various speeds are summarized in
Table 3.13. At 10 km/h, all the damped suspension configurations have better reduction in the PTP
values of BA than the undamped suspension. Also, the percent reduction is increased when the
suspension have more than 3 dampers. This appears from the table as the suspension
configurations S5-all and S4-1245 have the highest reduction in the PTP values.
It is noticed that when the speed is increased to 40 km/h, the suspension configuration S1-1 and
S5-all have the lowest BA peaks. At 60 km/h, all the suspension configurations have a poorer
performance than S0. All the suspension configurations except the suspension configuration S1-1
increase the PTP values of the BA. The RMS values of the BA responses are listed in Table 3.13.
The table shows that the number of the dampers affects the BA RMS and the suspension
configuration S1-1, S2-15 and S3-125 have the biggest reductions in the RMS values at all speeds.
Similarly, the AA responses under the medium excitation are shown in Figure 3.12(a, b). It is seen
that all the damped suspension configurations have a significant effect on the suspension
performance. The damped suspension improves the AA responses at all speeds. Figure 3.12(b)
illustrates the frequency domain of the AA responses under the same excitation. The figure shows
that the undamped AA response has a peak amplitude at 0.9 Hz at all speeds and that the damped
suspensions have reduced the AA resonant peaks at 0.9 Hz at all speeds. At speed of 10 km/h, the
peaks lie between the pitch frequency of 0.9 Hz and excitation frequency of 1.11 Hz. But at 40
km/h and 60 km/h, the peaks occur at frequencies between the pitch frequency of 0.9 Hz and the
excitation frequencies of 4.44 Hz and 6.66 Hz, respectively.
69
The PTP values of the AA are listed in Table 3.14. At 10 km/h, the damped suspension
performance has improved. In addition, the suspension configurations S4-1245 and S5-all offer the
highest reduction in the PTP values of AA. This is followed by the suspension configurations S1-1
and S2-15. For speeds 40 and 60 km/h, the undamped suspension has the lowest peak values.
Table 3.14 summarizes the RMS values of the AA responses. It is seen that increasing the number
of dampers at 10 and 40 km/h offer a significant effect on the suspension performance. However,
at 60 km/h, the dampers have a negative effect on the vehicle performance as the response is
deteriorated when the number of dampers is increased.
3- Sharp bump responses
In this section, the suspension performance is investigated under severe excitation which is the
sharp road profile with 0.5 m width (too short for full-size tracked vehicles). The bounce and pitch
acceleration of the undamped and damped suspension configurations are shown in Figure 3.13(a,
b) and Figure 3.14(a, b). The BA responses shown in Figure 3.13(a) indicate that under the sharp
terrains the suspension performance is highly influenced by the number of dampers fitted to the
suspension. The suspension responses are significantly increased when the suspension is fitted
with any dampers. Figure 3.13(b) shows the bounce peak amplitude of the undamped suspension
at frequency of 1.5 Hz at all speeds. At speed of 10 km/h, the damped suspension responses show
multiple peaks after the bounce frequency of 1.5 Hz due to the multiple suspension dynamics
discussed earlier in section 3.7.1.
When the vehicle traverses over the same sharp bump road at the medium speed of 40 km/h, the
S1-1 and S5-all suspension configurations have reduced the BA peaks while configurations S2-15,
S3-125 and S4-1245 increase the BA peaks as shown in and Figure 3.13(a). However, it is clear
from Figure 3.13(b) that this increase in the BA peaks is due to the wheel bounce of frequency 12.6
Hz. The hull bounce, which occurs around 4.5 Hz, is considerably suppressed in amplitude due to
the damped suspension compared to the undamped suspension case. Similarly, when the vehicle
traverses the sharp bump at the high speed of 60 km/h, the S1-1 and S3-125 configurations have
the highest reduction in the BA peaks. In addition, Figure 3.13(a, b) shows the peak values of the
BA responses in both the time and frequency domains are less than the corresponding BA
responses at the medium speed of 40 km/h.
Table 3.15 lists the PTP values of the BA responses of the suspension configurations under sharp
bump excitation. It can be seen that the BA responses are significantly increased when the
suspension is equipped with any number of dampers. The configurations S1-1 and S2-15 offer the
lowest PTP values at all speeds. This shows that the increase in the damping forces is not always
favourable to vibration attenuation especially over rough bump roads.
The time history of the AA of all suspension configurations under the same sharp bump road
excitation is illustrated in Figure 3.14(a). The response of the damped suspension configurations
have a lower settling time than the undamped suspension. However, the undamped suspension
PTP responses are lower than the damped suspension configurations. The frequency domain
responses shown in Figure 3.14(b) show that the undamped suspension has a pitch resonance
frequency of 0.9 Hz at all speeds. At speed of 10 km/h, the damped suspension has multiple peak
70
amplitudes between the pitch frequency of 0.9 Hz and the excitation frequency of 5.55 Hz. Also, at
all speeds, the bounce responses are not excited. Also, the figure indicates that the resonant AA
peaks of the damped suspension are reduced at all speeds.
The suspension configurations S1-1 and S2-15 have the lowest increase in the PTP values of the
AA amplitudes as can be seen from Table 3.16.The results listed in the table show the RMS values
of the AA under different speeds. The data show that all the damped suspension configurations
have reduced the RMS values of the AA responses at 10 and 40 km/h speeds. However, the
damped suspension is unable to suppress the vibration from the road at the wheel bounce
frequencies.
Over the sharp bump road, the damped suspension system is able to effectively isolate vehicle
vibration at hull bounce and pitch resonant frequencies but fails to effectively suppress the vibration
at wheel bounce frequencies. For shallow bump roads, the required damping force necessary to
damp the vibration is small. But when the vehicle traverses a rough road or increases its speed, the
dampers should have a sufficient damping force to suppress the road vibration. It can be seen that
increasing the number of dampers in the suspension offers a better suspension performance at low
speed under shallow roads. However, the suspension with the same configurations is less effective
in suppressing the road vibration for medium and sharp bump roads.
71
Speed 10 km/h
2 1.4
Undamped Undamped
At wh1 At wh1
1.5 At wh15 At wh15
1.2
At wh125 At wh125
At wh1245 At wh1245
1 At all wheels At all wheels
1
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 40 km/h
8 6
Undamped Undamped
At wh1 At wh1
6 At wh15 At wh15
At wh125 5 At wh125
At wh1245 At wh1245
4
2 4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0
3
-2
-4 2
-6
1
-8
-10 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
Speed 60 km/h
8 5
Undamped Undamped
At wh1 At wh1
6 4.5
At wh15 At wh15
At wh125 At wh125
At wh1245 4 At wh1245
4
3.5
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
3
0
2.5
-2
2
-4
1.5
-6
1
-8 0.5
-10 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
(a) (b)
Figure 3.11 Bounce acceleration responses under medium bump excitation
72
Speed 10 km/h
3 2.5
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
2 At wh125 At wh125
At wh1245 2 At wh1245
1

1.5
-1
0.5
-2
-3 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 40 km/h
5 0.8
Undamped Undamped
At wh1 At wh1
4
At wh15 0.7 At wh15
At wh125 At wh125
3 At wh1245 At wh1245
At all wheels 0.6 At all wheels
2

0.5
1
0 0.4
-1
0.3
-2
0.2
-3
0.1
-4
-5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.4
Undamped Undamped
At wh1 At wh1
At wh15 0.35 At wh15
4 At wh125 At wh125
At wh1245 At wh1245
At all wheels 0.3 At all wheels
2
0.25
0 0.2
0.15
-2
0.1
-4
0.05
-6 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.12 Angular acceleration responses under medium bump excitation
73
Speed 10 km/h
8 1.4
Undamped Undamped
At wh1 At wh1
6 At wh15 At wh15
1.2
At wh125 At wh125
At wh1245 At wh1245
1
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
-2
0.4
-4
0.2
-6
-8 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
Speed 40 km/h
10 3.5
Undamped Undamped
At wh1 At wh1
8 At wh15 At wh15
3
At wh125 At wh125
At wh1245 At wh1245
6
2.5
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
2
0 1.5
-2
1
-4
0.5
-6
-8 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
Speed 60 km/h
10 1.8
Undamped Undamped
At wh1 At wh1
8 At wh15 1.6 At wh15
At wh125 At wh125
At wh1245 At wh1245
1.4
1.2
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
1
2
0.8
0
0.6
-2
0.4
-4 0.2
-6 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.13 Bounce acceleration responses under sharp bump excitation
74
Speed 10 km/h
4 0.7
Undamped Undamped
At wh1 At wh1
3 At wh15 At wh15
0.6
At wh125 At wh125
At wh1245 At wh1245
0.5

1
0.4
0.3
-1
0.2
-2
0.1
-3
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
5 1.8
Undamped Undamped
At wh1 At wh1
4 1.6
At wh15 At wh15
At wh125 At wh125
3 At wh1245 At wh1245
1.4
2
1.2
1 Angular Acc (rad/s2)

1
0
0.8
-1
0.6
-2
0.4
-3
-4 0.2
-5 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
Speed 60 km/h
5 0.9
Undamped Undamped
At wh1 At wh1
4 0.8
At wh15 At wh15
At wh125 At wh125
3 At wh1245 At wh1245
0.7
2
0.6
1
0.5
0
0.4
-1
0.3
-2
0.2
-3
-4 0.1
-5 0
0 0.5 1 1.5 2 2.5 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.14 Angular acceleration responses under sharp bump excitation
75
Table ‎3.11 PTP and RMS values of bounce acceleration under shallow bump excitation
-2 -2
Description PTP (ms ) RMS (ms )
V=10 PTP% V=40 PTP% V=60 PTP% V=10 rms% V=40 rms% V=60 rms%
S0 0.78 0.0 19.69 0.0 18.88 0.0 0.15 0.0 6.55 0.0 6.35 0.0
S1-1 0.76 2.6 5.32 73.0 7.95 57.9 0.11 26.7 0.49 92.5 0.74 88.3
S2-15 0.59 24.4 9.00 54.3 12.98 31.2 0.09 40.0 0.82 87.5 1.08 83.0
S3-125 0.56 28.2 6.31 68.0 11.42 39.5 0.09 40.0 0.58 91.1 0.92 85.5
S4-1245 0.53 32.1 7.84 60.2 14.59 22.7 0.08 46.7 0.69 89.5 1.16 81.7
S5-all 0.49 37.7 8.49 56.9 18.14 3.9 0.08 46.7 0.75 88.5 1.34 78.9
Table ‎3.12 PTP and RMS values of angular acceleration under shallow bump excitation
-2 -2
Description PTP (rads ) RMS (rads )
S0 2.61 0.0 4.14 0.0 3.28 0.0 0.74 0.0 1.15 0.0 0.60 0.0
S1-1 1.32 49.4 5.33 -28.7 6.10 -85.8 0.19 74.3 0.53 53.9 0.51 14.5
S2-15 1.22 53.3 6.75 -63.0 7.91 -141.0 0.18 75.7 0.56 51.3 0.56 5.5
S3-125 1.27 51.3 7.8 -88.4 9.89 -201.2 0.19 74.3 0.65 43.5 0.70 -17.1
S4-1245 1.2 54.0 8.19 -97.8 10.31 -214.1 0.18 75.7 0.66 42.6 0.71 -18.9
S5-all 1.2 54.0 8.21 -98.3 10.37 -215.8 0.18 75.7 0.66 42.6 0.71 -19.3
76
Table ‎3.13 PTP and RMS values of bounce acceleration under medium bump excitation
-2 -2
PTP (ms ) RMS (ms )
Description
S0 3.28 0.0 12.40 0.0 10.73 0.0 0.97 0.0 4.14 0.0 3.54 0.0
S1-1 2.82 13.8 9.16 26.1 10.65 0.8 0.35 63.8 0.55 86.8 0.59 83.3
S2-15 2.31 29.5 11.37 8.3 14.55 -35.5 0.31 68.3 0.91 77.9 0.93 73.8
S3-125 2.47 24.6 13.86 -11.8 16.85 -57.0 0.31 68.4 0.99 76.1 0.96 72.8
S4-1245 1.98 39.4 11.98 3.4 16.42 -53.0 0.23 76.1 0.97 76.6 1.05 70.4
S5-all 1.05 68.0 9.53 23.1 14.72 -37.1 0.14 85.9 0.74 82.2 1.04 70.7
Table ‎3.14 PTP and RMS values of angular acceleration under medium bump excitation
-2 -2
PTP (rads ) RMS (rads )
Description
S0 5.76 0.0 3.16 0.0 3.26 0.0 1.85 0.0 0.65 0.0 0.33 0.0
S1-1 1.99 65.4 5.26 -66.8 6.47 -98.3 0.32 82.9 0.40 37.5 0.39 -18.1
S2-15 2.24 61.1 6.46 -104.5 7.49 -129.6 0.30 83.9 0.50 22.6 0.49 -46.8
S3-125 2.36 59.1 7.32 -132.1 8.60 -163.4 0.30 83.6 0.57 11.7 0.59 -78.8
S4-1245 1.96 66.0 8.11 -157.0 9.82 -200.8 0.26 85.9 0.62 3.9 0.64 -92.1
S5-all 1.95 66.1 8.13 -157.5 9.88 -202.7 0.26 85.9 0.62 3.5 0.64 -92.9
77
Table ‎3.15 PTP and RMS values of bounce acceleration under sharp bump excitation
-2 -2
S0 5.19 0.0 6.48 0.0 6.10 0.0 1.23 0.0 1.58 0.0 1.53 0.0
S1-1 10.69 -105.8 13.34 -105.8 12.03 -97.2 1.03 16.4 0.65 58.9 0.35 77.0
S2-15 13.21 -154.4 15.76 -143.0 14.14 -131.8 1.15 6.4 2.31 -46.3 1.42 6.6
S3-125 14.60 -181.2 14.54 -124.2 13.54 -122.1 1.24 -1.1 2.73 -72.8 0.57 62.9
S4-1245 14.76 -184.2 15.46 -138.4 13.74 -125.3 1.39 -13.2 1.88 -18.8 1.44 5.7
S5-all 14.39 -177.0 14.30 -120.6 13.68 -124.3 1.59 -29.5 0.77 51.4 0.59 61.5
Table ‎3.16 PTP and RMS values of angular acceleration under sharp bump excitation
-2 -2
S0 2.29 0.0 4.16 0.0 2.09 0.0 0.60 0.0 1.22 0.0 0.12 0.0
S1-1 5.20 -126.8 7.02 -69.0 6.36 -203.8 0.31 48.4 1.45 -19.0 0.72 -495.5
S2-15 6.38 -178.6 8.18 -96.7 9.54 -355.5 0.44 26.6 0.97 20.6 0.27 -120.2
S3-125 6.53 -184.7 8.62 -107.3 9.26 -342.3 0.47 22.3 0.58 52.0 0.45 -269.0
S4-1245 6.43 -180.5 8.78 -111.1 8.81 -320.5 0.48 20.0 0.34 72.3 0.27 -125.1
S5-all 6.42 -180.3 8.76 -110.8 8.81 -320.9 0.48 20.1 0.34 72.3 0.27 -125.3
78
3.7.4 Influence of damping coefficient
In this section, the effect of varying the damping coefficient of the passive damper on the vehicle
performance is studied. The equivalent damping coefficient listed in the table is modified to yield
the low and high damping force. This is achieved by increasing or reducing the damping value by ±
20% of the nominal value. Thus, three damping values namely: nominal, low and high damping are
used in the simulation. This is equivalent to 22,520 Ns/m, 18,016 Ns/m and 27,024 Ns/m
respectively.
The influence of the variations in the suspension damping is shown in Figure 2.12(a, b) and
Figure 3.16(a, b). The BA responses in Figure 3.15(a) reveals that an increase in the suspension
damping offers a reduction in the PTP amplitudes at low speed of 10 km/h. However, when the
vehicle speed is further increased to 40 km/h and 60 km/h, the suspension performance is reduced
and the PTP values are increased as shown in Table 3.17. Moreover, the BA spectra presented in
Figure 3.15(b) show pitch and bounce peak amplitudes at frequencies of 0.9 Hz and 1.5 Hz as well
as peak amplitude at excitation frequency of 0.55 Hz at speed of 10 km/h. The other peaks seen in
the FFT plot are almost harmonics of the excitation frequency. But the frequencies of the first two
peaks are modulated by the pitch and bounce frequencies. The extra peaks are due to the tracked
vehicle dynamics. For speeds of 40 km/h and 60 km/h, the suspension response show peak
amplitudes between the bounce frequency of 1.5 Hz and the excitation frequencies of 2.22 Hz and
3.33 Hz. Also, the figure demonstrate that the increase in the suspension damping offers better
isolation at the neighbourhood of bounce resonant frequency. This improvement in the suspension
is achieved at low speed of 10 km/h. However, increasing the speed over 2 Hz i.e. at a speed of 60
km/h increases the RMS amplitudes and deteriorates the suspension performance as seen from
Table 3.17.
the same bump excitation and speeds. The AA responses shown in Figure 3.16(a ) show that as
the damping increased, the suspension performance is improved at low speed. However, at
medium and high speeds the suspension performance deteriorates due to high damping levels.
This appears in the PTP values of AA shown in Table 3.17. The FFT values of the AA responses
show peak amplitudes between the excitation frequency of 0.5 Hz and bounce frequency of 1.5 Hz
at a speed of 10 km/h. At the high speeds of 40 km/h and 60 km/h, the suspension responses
show peaks at the excitation frequencies of 2.22 Hz and 3.33 Hz respectively. In addition, the
figure shows the same trend as the BA responses where the increase in the damping offers good
isolation at low frequency range (below 2 Hz). Although, the suspension performance is degraded
and the RMS values are increased when the frequency exceeded 2 Hz as shown in Table 3.17.
Over the medium bump road, the excitation frequency is increased and the suspension is
subjected to high vibration level as a result. Figure 3.17(a, b) show the BA of the system responses
in time and frequency domains. The time history of the BA shown in Figure 3.17(a) demonstrates
79
that increasing the suspension damping degrades the suspension performance for all vehicle
speeds. The suspension with low damping offer smooth ride of the vehicle but with large settling
time. The BA spectra shown in Figure 3.17(b) shows the bounce peak amplitude of the suspension
at the natural frequency of 1.5 Hz. At a speed of 10 km/h, the suspension response shows peak
amplitudes at pitch and bounce frequencies of 0.9 Hz and 1.5 Hz as well as multiple peaks due to
the tracked vehicle dynamics discussed in section 3.7.1. at speeds of 40 km/h and 60 km/h, the
suspension responses show peaks between the bounce frequency of 1.5 Hz and excitation
frequency of 4.44 Hz and 6.66 Hz respectively. Also, the figure shows good isolation in the vicinity
of the bounce resonant frequency. However, the suspension behaviour is reduced by increasing
the excitation frequency greater than 2 Hz. This can be noticed from the results presented in
Table 3.18 which summarizes the PTP and RMS values of the BA responses.
the same bump excitation and speeds. The time history of the AA responses shows that as the
damping increased, the suspension performance is reduced at all speeds. This appears in the PTP
values of AA shown in Table 3.17. The FFT values of the AA responses at a speed of 10 km/h
show peak amplitudes at pitch frequency of 0.9 Hz and excitation frequency of 1.11 Hz. At high
speeds of 40 km/h and 60 km/h, the peak amplitudes lie between the pitch frequency and the
excitation frequencies of 4.44 Hz and 6.66 Hz. Also, the figure shows that increasing the damping
degrades the suspension performance. The peak to peak (PTP) and RMS values of the AA
responses at different speeds are listed in Table 3.18
In this section, the suspension is investigated under sever bump excitation. The BA and AA
responses of the suspension system under the sharp excitation are shown in Figure 3.19(a, b) and
Figure 3.20(a, b). The BA time responses indicate that the increase in the suspension yields a
significant increase in the suspension responses. The BA spectra shown in Figure 3.19(b) show
that at low speed of 10 km/h, the suspension has a bounce peak amplitude at frequency of 1.5 Hz
as well as multiple peaks due to the suspension dynamics. However, the pitch response is not
excited. At speeds of 40 km/h and 60 km/h, both the pitch and the bounce peak amplitude
responses are not excited because the excitation frequencies lie between 22.22 Hz and 33.33 Hz.
Also, the figure indicates that the suspension has a significant increase around the wheel resonant
frequency of about 12.5 Hz. However, the increase in the damping force increases the peak
amplitude in the frequency range up to 10 Hz. Also, around the wheel resonant frequency, the
increase in the damping does not indicate any significant change. The PTP values of the BA listed
in Table 3.19 confirm that the PTP values are significantly increased by increasing the damping for
all speeds.
Similarly, the AA responses under the same excitation are shown in Figure 3.20(a, b). The AA
responses shown in Figure 3.20(a) indicate that the suspension response is increased by
increasing the damping force. Furthermore, Figure 3.20(b) show the FFT responses of the AA
which demonstrate the pitch and bounce peak amplitudes at frequencies of 0.9 Hz and 1.5 Hz,
80
respectively, in addition to the multiple peaks at a speed of 10 km/h due to the suspension
dynamics. At the higher speeds of 40 km/h and 60 km/h, both the pitch and the bounce peak
amplitude responses are not excited because the excitation frequencies lie between 22.22 Hz and
33.33 Hz. Furthermore, increasing the damping offers no significant improvement in the
suspension performance. The RMS values listed in Table 3.19 indicate that the AA responses are
increased when the damping is increased.
Table ‎3.17 Influence of variation of damping on PTP and RMS values under shallow bump
Vehicle speed (km/h)
Response Damping
V=10 V=40 V=60
(Ns/m)
PTP RMS PTP RMS PTP RMS
L 0.582 0.0018 5.942 0.0120 10.173 0.0179
2
BA (m/s ) N 0.564 0.0018 6.310 0.0116 11.418 0.0184
H 0.546 0.0018 6.987 0.0118 12.633 0.0192
L 1.332 0.0040 6.783 0.0114 8.222 0.0119
AA
2 N 1.267 0.0037 7.801 0.0129 9.891 0.0139
(rad/s )
H 1.221 0.0036 8.773 0.0144 11.430 0.0159
Table ‎3.18 Influence of variation of damping on PTP and RMS values under medium bump
Response Damping
V=10 V=40 V=60
(Ns/m)
L 2.461 0.0061 11.599 0.0173 14.825 0.0179
2
BA (m/s ) N 2.471 0.0061 13.861 0.0198 16.856 0.0192
H 2.484 0.0062 16.341 0.0223 18.567 0.0204
L 2.343 0.0063 6.040 0.0099 7.668 0.0103
AA
2 N 2.356 0.0061 7.326 0.0114 8.598 0.0119
(rad/s )
H 2.402 0.0060 8.465 0.0128 9.717 0.0133
Table ‎3.19 Influence of variation of damping on PTP and RMS values under sharp bump
Response Damping
V=10 V=40 V=60
(Ns/m)
L 12.503 0.0216 13.857 0.0527 12.836 0.0108
2
BA (m/s ) N 14.609 0.0248 14.569 0.0529 13.588 0.0113
H 16.477 0.0277 15.111 0.0530 14.061 0.0117
L 5.714 0.0082 8.216 0.0112 8.652 0.0086
AA
2 N 6.526 0.0094 8.639 0.0114 9.296 0.0088
(rad/s )
H 7.190 0.0104 8.927 0.0115 9.773 0.0090
L:18016; N: : 22520; H: 27024
81
Speed 10 km/h
0.3 0.06
Nominal: 22520 Ns/m Nominal: 22520 Ns/m
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m High: 27024 Ns/m
0.2
0.05
0.1
0.04
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0
0.03
-0.1
0.02
-0.2
0.01
-0.3
-0.4 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.35
Low: 18016 Ns/m Low: 18016 Ns/m
3 High: 27024 Ns/m High: 27024 Ns/m
0.3
2
0.25
1
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
0.15
-1
0.1
-2
0.05
-3
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.35
Low: 18016 Ns/m Low: 18016 Ns/m
4 0.3
2 0.25
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 0.2
-2 0.15
-4 0.1
-6 0.05
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.15 Influence of varying damping coefficient on BA responses under shallow bump
82
Speed 10 km/h
0.8 0.16
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 0.14 High: 27024 Ns/m
0.6
0.12
0.4

0.1
0.2
0.08
0
0.06
-0.2
0.04
-0.4
0.02
-0.6 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
5 0.35
Low: 18016 Ns/m Low: 18016 Ns/m
4
0.3
3
2 0.25

1
0.2
0.15
-1
-2 0.1
-3
0.05
-4
-5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.25
Low: 18016 Ns/m Low: 18016 Ns/m
4
0.2
2
0.15
0
-2
0.1
-4
0.05
-6
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.16 Influence of varying damping coefficient on AA responses under shallow bump
83
Speed 10 km/h
1.5 0.2
Low: 18016 Ns/m Low: 18016 Ns/m
0.18
1
0.16
0.14
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.12
0 0.1
0.08
-0.5
0.06
0.04
-1
0.02
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
8 0.35
Low: 18016 Ns/m Low: 18016 Ns/m
0.3
4
0.25
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
0
-2 0.15
-4
0.1
-6
0.05
-8
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
10 0.25
Low: 18016 Ns/m Low: 18016 Ns/m
5 0.2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 0.15
-5 0.1
-10 0.05
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.17 Influence of varying damping coefficient on BA responses under medium bump
84
Speed 10 km/h
1.5 0.25
Low: 18016 Ns/m Low: 18016 Ns/m
1
0.2
0.5

0.15
0.1
-0.5
0.05
-1
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
5 0.25
Low: 18016 Ns/m Low: 18016 Ns/m
4
3 0.2
2

1 0.15
-1 0.1
-2
-3 0.05
-4
-5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.18
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 0.16 High: 27024 Ns/m
4
0.14
2 0.12
0.1
0
0.08
-2 0.06
0.04
-4
0.02
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.18 Influence of varying damping coefficient on AA responses under medium bump
85
Speed 10 km/h
10 0.7
Low: 18016 Ns/m Low: 18016 Ns/m
8
0.6
6
4 0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2
0.4
0.3
-2
-4 0.2
-6
0.1
-8
-10 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 3.5
Low: 18016 Ns/m Low: 18016 Ns/m
3
6
2.5
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
2
0 1.5
-2
1
-4
0.5
-6
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
10 0.18
Low: 18016 Ns/m Low: 18016 Ns/m
8 High: 27024 Ns/m 0.16 High: 27024 Ns/m
0.14
6
0.12
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.1
2
0.08
0
0.06
-2
0.04
-4 0.02
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.19 Influence of varying damping coefficient on BA responses under sharp bump
86
Speed 10 km/h
4 0.18
Low: 18016 Ns/m Low: 18016 Ns/m
3 High: 27024 Ns/m 0.16 High: 27024 Ns/m
0.14
2
0.12

1
0.1
0
0.08
-1
0.06
-2
0.04
-3 0.02
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
5 0.7
Low: 18016 Ns/m Low: 18016 Ns/m
4
0.6
3
2 0.5
1 Angular Acc (rad/s2)

0.4
0.3
-1
-2 0.2
-3
0.1
-4
-5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
5 0.5
Low: 18016 Ns/m Low: 18016 Ns/m
4 0.45
3 0.4
2 0.35
1 0.3
0 0.25
-1 0.2
-2 0.15
-3 0.1
-4 0.05
-5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.20 Influence of varying damping coefficient on AA responses under sharp bump
87
3.7.5 Influence of suspension stiffness
The torsion bar represents the elastic element in the tracked vehicle suspension system which
connects the road wheels to the vehicle hull. Thus, the equivalent suspension stiffness is
represented in the simulation by the dependence of the vertical force on a road wheel with the
resultant vertical displacement of that wheel. The nominal value of the suspension stiffness is
104,000 N/m (Table 3.1). The effect of the suspension stiffness on the vehicle dynamic
performance is studied by choosing low and high values of the nominal suspension stiffness which
are 83,200 N/m and 124,800 N/m.
Figure 3.21 (a, b) demonstrates the time and frequency responses of the BA of the vehicle that
negotiates a shallow bump road profile with different speeds. As shown in the figure, the higher
suspension stiffness yields reasonably bigger acceleration levels. This is due to the higher energy
levels that are transmitted to the hull mass. The absolute values of the FFT responses of the BA
are also shown in Figure 3.21(b). the figure shows the peak amplitudes at excitation frequency of
0.55 Hz and pitch and bounce frequencies of 0,9 Hz and 1.5Hz at low speed of 10 km/h. The other
peaks seen in the FFT plot are due to the tracked vehicle dynamics discussed in section 3.7.1. At
speeds of 40 km/h and 60 km/h, the suspension has peak amplitudes that lie between the bounce
frequency of 1.5 Hz and excitation frequencies of 2.22 Hz and 3.33 Hz. The BA responses are
decreased in the bounce resonant frequency by increasing the suspension stiffness. Moreover, the
wheel bounce resonant frequencies are shifted up when the suspension has higher stiffness.
The PTP values of the BA responses with different values of suspension stiffness at various
speeds are listed in Table 3.20. It is seen that at low speed, increasing the suspension stiffness
enhances the suspension performance and reduces the peak amplitude. However, the suspension
responses decrease when the suspension has a high stiffness. The RMS values of the system
responses are listed in Table 3.20. A significant increase in the RMS values of the system
response occurs at the bounce resonant frequency at medium and high speed. However, at low
speed, high suspension stiffness gives a relatively smooth ride comfort.
the same bump excitation. The results shown in Figure 3.22(a) demonstrates that softer
suspension yields a reduction in the AA responses at all speeds. In Figure 3.22(b), the suspension
response shows pitch peak amplitude at the natural of frequency of 0.9 Hz at the low speed of 10
km/h. In addition, the peak amplitudes of the suspension at 40 km/h and 60 km/h lie between the
pitch frequency of 0.9 Hz and the excitation frequencies of 2.22 Hz and 3.33 Hz. The variation in
the suspension stiffness affect the pitch and wheel resonant frequencies. The stiffer suspension
increase the peak amplitude in the vicinity of the pitch resonant frequency at all speeds. Also, the
softer suspension stiffness yields lower bounce resonant frequency for the wheels. Table 3.20
shows the PTP values of the AA responses for different suspension stiffness levels. The results
show the increase in the PTP values due suspension stiffness increase. The RMS values of the AA
responses under the shallow bump excitation are summarized in Table 3.20.
88
The influence of the variation in the suspension stiffness on the vehicle performance is investigated
over medium bump profile. Figure 3.23(a, b) show the BA of the suspension responses in time and
frequency domains. The time history of the BA shown in Figure 3.23(a) demonstrates that the
stiffer suspension yields to smooth acceleration ride at the selected speeds. Conversely, the BA
spectra shown in Figure 3.23(b) show the pitch and bounce peak amplitudes at frequencies of 0.9
Hz and 1.5 Hz in addition to other peaks due to the suspension behaviour. For speeds of 40 km/h
and 60 km/h, the peak amplitudes occur at excitation frequencies of 4.44 Hz and 6.66 Hz.
Moreover, the suspension response indicates slight increase in the peak amplitude at bounce
resonant frequency when the suspension stiffness is increased. Table 3.21 shows that the increase
in the suspension stiffness offers low PTP values. The RMS. values of the BA listed in Table 3.21
show the reduction in the peak amplitude of the suspension responses at frequencies greater than
2 Hz when the suspension stiffness is increased.
In addition, the AA responses under the same excitation are presented in Figure 3.24(a, b). it is
shown from Figure 3.24(a) that the AA responses are increased when the stiffness is increased.
The FFT values of the AA responses shown in Figure 3.24(b) show two dominant peak amplitudes
at the pitch and bounce frequencies at low speed. The minor peaks are due to multiple excitations
of the wheels of the tracked vehicle. In addition, the peak amplitudes of the suspension at 40 km/h
and 60 km/h are due to the excitation frequencies of 4.44 Hz and 6,66 Hz. Also, the figure reveals
that as the stiffness is increased, the peak amplitude is also increased especially at the
neighbourhood of the pitch resonant frequency (0.9 Hz). As the vehicle speed is increased, the
PTP values of the AA are also increased as seen from Table 3.21.
The change in the suspension stiffness and its effect on the vehicle performance is further
investigated under sharp bump excitation. The time and frequency domains of the BA responses
are shown in Figure 3.25(a, b). The BA responses in Figure 3.25(a) show the significant increase in
the suspension response due to high suspension stiffness. This is due to the high energy
transmitted to the vehicle hull. The FFT amplitudes of the BA show bounce peak amplitude at the
bounce frequency of 1.5 Hz in addition to multiple peaks at low speed of 10 km/h. The figure also
shows no change in the peak amplitudes around the bounce resonant frequency. However, there is
a significant increase in the peak amplitude at the wheel resonant frequency. The increase in the
BA responses can be noticed from the increased in the PTP values as shown in Table 3.22.
Similarly, the AA responses under the same excitation are shown in Figure 3.26(a, b). The time
history shown Figure 3.26(a) show the improvement in the suspension responses at low and
medium speeds due to the increase in the suspension stiffness. However, the suspension
responses are increased at speed 60 km/h when the suspension is further increased. In addition,
Figure 3.26(b) show the AA spectra which shows that at low speed, there is a slight increase in the
peak amplitude around 2 Hz when the stiffness is increased. However, there is a significant
89
increase in the peak amplitudes at the wheel resonant frequency at medium and high speeds.
Table 3.22 summarizes the RMS values of the AA responses at different speeds.
Speed 10 km/h
0.4 0.06
Nominal: 104000 N/m Nominal: 104000 N/m
Low: 83200 N/m Low: 83200 N/m
High: 124800 N/m High: 124800 N/m
0.3
0.05
0.2
0.04
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.1
0.03
0.02
-0.1
0.01
-0.2
-0.3 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.25
Low: 83200 N/m Low: 83200 N/m
3 High: 124800 N/m High: 124800 N/m
0.2
2
1
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.15
0.1
-1
-2
0.05
-3
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.4
Low: 83200 N/m Low: 83200 N/m
High: 124800 N/m 0.35 High: 124800 N/m
4
0.3
2
0.25
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
-2
0.15
-4
0.1
-6
0.05
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.21 Influence of suspension stiffness on BA responses under shallow bump
90
Speed 10 km/h
0.8 0.14
Low: 83200 N/m Low: 83200 N/m
High: 124800 N/m High: 124800 N/m
0.6 0.12
0.4 0.1

0.2 0.08
0 0.06
-0.2 0.04
-0.4 0.02
-0.6 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.35
Low: 83200 N/m Low: 83200 N/m
3 High: 124800 N/m High: 124800 N/m
0.3
2
0.25
1

0.2
0
-1 0.15
-2
0.1
-3
0.05
-4
-5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
4 0.25
Low: 83200 N/m Low: 83200 N/m
High: 124800 N/m High: 124800 N/m
2
0.2
0
0.15
-2
0.1
-4
0.05
-6
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.22 Influence of suspension stiffness on AA responses under shallow bump
91
Table ‎3.20 Influence of variation of suspension stiffness on PTP and RMS values under shallow
bump at different speed
Suspension Vehicle speed (km/h)
Response stiffness V=10 V=40 V=60
(N/m) PTP RMS PTP RMS PTP RMS
L 0.568 0.00182 6.258 0.0110 11.115 0.0170
2
BA (m/s ) N 0.564 0.00177 6.310 0.0116 11.418 0.0184
H 0.551 0.00173 6.424 0.0123 11.804 0.0200
L 1.230 0.00365 7.387 0.0122 9.728 0.0136
AA
2 N 1.267 0.00373 7.801 0.0129 9.891 0.0139
(rad/s )
H 1.278 0.00374 8.242 0.0137 10.073 0.0143
Table ‎3.21 Influence of variation of suspension stiffness on PTP and RMS values under medium
L 2.540 0.0064 14.252 0.0199 17.142 0.0191
2
BA (m/s ) N 2.471 0.0061 13.861 0.0198 16.856 0.0192
H 2.352 0.0058 13.517 0.0197 16.613 0.0195
L 2.209 0.0057 7.266 0.0110 8.509 0.0117
AA
2 N 2.356 0.0061 7.325 0.0114 8.598 0.0119
(rad/s )
H 2.437 0.0063 7.413 0.0119 8.803 0.0120
Table ‎3.22 Influence of variation of suspension stiffness on PTP and RMS values under sharp
L 14.426 0.02399 14.505 0.0428 13.583 0.0110
2
BA (m/s ) N 14.609 0.02477 14.569 0.0529 13.588 0.0113
H 14.809 0.02552 14.636 0.0632 13.613 0.0119
L 6.584 0.00937 8.693 0.0099 9.281 0.0078
AA
2 N 6.526 0.00938 8.639 0.0114 9.296 0.0088
(rad/s )
H 6.481 0.00941 8.596 0.0131 9.303 0.0098
L:83200 N/m; N: : 104000 N/m; H: 124800 N/m
92
Speed 10 km/h
1.5 0.25
Low: 83200 N/m Low: 83200 N/m
High: 124800 N/m High: 124800 N/m
1
0.2
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.15
0.1
-0.5
0.05
-1
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
6 0.35
Low: 83200 N/m Low: 83200 N/m
4 High: 124800 N/m High: 124800 N/m
0.3
2
0.25
0
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.2
-2
0.15
-4
0.1
-6
0.05
-8
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
8 0.25
Low: 83200 N/m Low: 83200 N/m
6 High: 124800 N/m High: 124800 N/m
0.2
4
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.15
0
-2
0.1
-4
-6
0.05
-8
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.23 Influence of suspension stiffness on BA responses under medium bump
93
Speed 10 km/h
1.5 0.2
Low: 83200 N/m Low: 83200 N/m
0.18
High: 124800 N/m High: 124800 N/m
1
0.16
0.14
0.5

0.12
0 0.1
0.08
-0.5
0.06
0.04
-1
0.02
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.2
Low: 83200 N/m Low: 83200 N/m
0.18
3 High: 124800 N/m High: 124800 N/m
0.16
2
0.14
1
0.12
0 0.1
0.08
-1
0.06
-2
0.04
-3
0.02
-4 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
4 0.16
Low: 83200 N/m Low: 83200 N/m
3
High: 124800 N/m 0.14 High: 124800 N/m
2
0.12
1
0.1
0
-1 0.08
-2
0.06
-3
0.04
-4
0.02
-5
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.24 Influence of suspension stiffness on AA responses under medium bump
94
Speed 10 km/h
8 0.7
Low: 83200 N/m Low: 83200 N/m
6 High: 124800 N/m High: 124800 N/m
0.6
4
0.5
Bounce Acc (m/s2) 2
Bounce Acc (m/s2)

0.4
0.3
-2
0.2
-4
0.1
-6
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 4
Low: 83200 N/m Low: 83200 N/m
8 High: 124800 N/m 3.5 High: 124800 N/m
6
3
4
2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2
2
0
1.5
-2
1
-4
-6 0.5
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
10 0.35
Low: 83200 N/m Low: 83200 N/m
8 High: 124800 N/m High: 124800 N/m
0.3
6
0.25
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
0.15
0
0.1
-2
0.05
-4
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.25 Influence of suspension stiffness on BA responses under sharp bump
95
Speed 10 km/h
4 0.16
Low: 83200 N/m Low: 83200 N/m
3 High: 124800 N/m 0.14 High: 124800 N/m
2 0.12

1 0.1
0 0.08
-1 0.06
-2 0.04
-3 0.02
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
5 0.8
Low: 83200 N/m Low: 83200 N/m
4 High: 124800 N/m 0.7 High: 124800 N/m
3
0.6
2
0.5
1
0.4
0
0.3
-1
0.2
-2
-3 0.1
-4 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
5 0.7
Low: 83200 N/m Low: 83200 N/m
4
High: 124800 N/m High: 124800 N/m
0.6
3
2 0.5
1
0.4
0.3
-1
-2 0.2
-3
0.1
-4
-5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.26 Influence of suspension stiffness on AA responses under sharp bump
96
3.7.6 Influence of wheel stiffness
The wheel stiffness is represented in the model with a stiffer spring due to the high rigidity of the
tracked vehicle road wheels. The influence of varying the wheel stiffness on the vehicle
performance is studied by choosing low and high stiffness from the nominal stiffness (613 kN/m)
which are 490.4 kN/m and 735.6 kN/m. The same excitation of three different bump road profiles is
used in this investigation.
Figure 3.27(a, b) shows the time and frequency responses of the BA of the vehicle that crosses a
shallow bump road with various speeds. The figure shows minor changes in the time history of the
BA responses at low speed when the wheel stiffness is increased. On the other hand, the BA
amplitudes are reduced by increasing the stiffness at 40 km/h and 60 km/h and the PTP values are
listed in Table 3.23. The FFT responses shown in Figure 3.27(b)show that the suspension
response at a speed of 10 km/h has peak amplitudes at excitation frequency of 0.55 Hz and pitch
and bounce amplitudes at frequencies of 0.9 Hz and 1.5 Hz. The other peak seen in the FFT plot
are due to the tracked vehicle dynamics. Also, the suspension response at 40 km/h and 60 km/h
has peak amplitudes at frequency of 2.22 Hz and 3.33 Hz, which are the excitation frequencies,
respectively. Furthermore, the responses reveal the increase in the peak amplitude due to the
increase in the wheel stiffness for frequencies greater than 4 Hz as can be seen from Table 3.23.
the same excitation. The results shown in Figure 3.28(a) demonstrate the increase in the AA
amplitudes when the stiffness is increased. However, increasing the stiffness reduce the AA
amplitudes at 10 km/h. The FFT values of the AA shown in Figure 3.28(b) show the pitch peak
amplitude at the pitch frequency of 0.9 Hz at the low speed of 10 km/h. In addition, at speeds of 40
km/h and 60 km/h the suspension has peak amplitudes at excitation frequencies of 2.22 Hz and
3.33 Hz. Also, the figure indicates that the AA peak amplitude is reduced when the stiffness is
increased and the RMS values of the AA responses are presented in Table 3.23.
Table ‎3.23 Influence of variation of wheel stiffness on PTP and RMS values at different speeds at
shallow bump
Wheel Vehicle speed (km/h)
L 0.566 0.00178 6.508 0.0121 11.430 0.0195
2
BA (m/s ) N 0.564 0.00177 6.310 0.0116 11.418 0.0184
H 0.560 0.00177 6.169 0.0114 11.314 0.0177
L 1.293 0.00383 7.623 0.0127 9.184 0.0132
2
AA (rad/s ) N 1.267 0.00373 7.801 0.0129 9.891 0.0139
H 1.251 0.00367 7.862 0.0130 10.298 0.0144
L: 490400 N/m; N: : 613000 N/m; H: 735600 N/m
over the medium bump profile. Figure 3.29(a, b) show the BA of the suspension responses in time
and frequency domains. The time history of the BA shown in Figure 3.28(a) demonstrates that the
97
increase in the wheel stiffness gives smooth acceleration ride at speed 10 km/h. However, when
the speed is further increased, the BA amplitudes are increased. Conversely, the BA spectra
shown in Figure 3.29(b) indicate significant increase in the peak amplitude at frequencies greater
than 2 Hz when the wheel stiffness is increased. At speed of 10 km/h, the figure shows peak
amplitude at the bounce frequency of 1.5 Hz. The other peaks are due to the tracked vehicle
dynamics. For speeds of 40 km/h and 60 km/h, the peak amplitudes are due to the excitation
frequencies of 4.44 Hz and 6.66 Hz. Table 3.24 shows the PTP values of the BA responses at
various speeds. The RMS. values of the BA responses at various speeds are listed in Table 3.24.
In addition, the AA responses under the same excitation are presented in Figure 3.30(a, b). It is
shown from Figure 3.30(a) that the AA responses are reduced at speed 10 km/h when the wheel
stiffness is increased.. The FFT values of the AA responses shown in Figure 3.30(b) show the peak
amplitudes at pitch frequency of 0.9 and excitation frequency of 1.11 Hz at speed of 10 km/h. For
speeds of 40 km/h and 60 km/h, the peak amplitudes are due to the excitation frequencies of 4.44
Hz and 6.66 Hz. Also, the figure reveals that as the stiffness is increased, the peak amplitude is
also increased at frequencies greater than 2 Hz. As the vehicle speed is further increased, the PTP
values of the AA are increased as seen from Table 3.24
medium bump
L 2.513 0.00617 12.277 0.0179 14.543 0.0172
2
BA (m/s ) N 2.471 0.00613 13.861 0.0198 16.856 0.0192
H 2.463 0.00610 15.525 0.0213 18.620 0.0213
L 2.382 0.00620 6.561 0.0110 8.138 0.0111
2
AA (rad/s ) N 2.356 0.00606 7.325 0.0114 8.598 0.0119
H 2.335 0.00596 7.817 0.0117 9.160 0.0125
L:490400 N/m; N: : 613000 N/m; H: 735600 N/m
The change in the wheel stiffness and its effect on the vehicle performance is further investigated
under sharp bump excitation. The time and frequency domains of the BA responses are shown in
Figure 3.31(a, b). The BA responses in Figure 3.31(a) show the significant increase in the BA
suspension response due to high wheel stiffness. This is due to the high rigidity of the road wheels
which transmits the shocks to the vehicle hull. The FFT plots of the BA show a peak amplitude at
the bounce frequency of 1.5 Hz as well as multiple peak amplitudes at the low speed of 10 km/h
due to the suspension dynamics. For speeds of 40 km/h and 60 km/h, the bounce response is not
excited because the excitation frequencies lie between 22.22 Hz and 33.33 Hz. However, there is a
significant increase in the peak amplitude at the wheel resonant frequency at all speeds. The
increase in the BA responses can be noticed from the increase in the PTP values as shown in
Table 3.25.
Similarly, the AA responses under the same excitation are shown in Figure 3.32 (a, b). The time
history shown Figure 3.32(a) show the increase in the suspension responses at all speeds due to
98
the increase in the suspension stiffness. In addition, Figure 3.32(b) show other peak amplitudes at
low speed due to the tracked vehicle dynamics. The AA spectra which show increase in the peak
amplitude at low speed around 2 Hz when the stiffness is increased. However, there is negligible
increase in the peak amplitude around the bounce and pitch resonant frequencies at medium and
high speeds. Table 3.25 summarizes the RMS values of the AA responses at different speeds.
99
Speed 10 km/h
0.3 0.06
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
0.2
0.05
0.1
0.04
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0
0.03
-0.1
0.02
-0.2
0.01
-0.3
-0.4 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.25
Low: 490400 N/m Low: 490400 N/m
3 High: 735600 N/m High: 735600 N/m
0.2
2
1
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.15
0.1
-1
-2
0.05
-3
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.4
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m 0.35 High: 735600 N/m
4
0.3
2
0.25
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
-2
0.15
-4
0.1
-6
0.05
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.27 Influence of wheel stiffness on BA responses under shallow bump
100
Speed 10 km/h
0.8 0.16
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m 0.14 High: 735600 N/m
0.6
0.12
0.4

0.1
0.2
0.08
0
0.06
-0.2
0.04
-0.4
0.02
-0.6 0
0 2 4 6 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.25
Low: 490400 N/m Low: 490400 N/m
3 High: 735600 N/m High: 735600 N/m
0.2
2
1

0.15
0
-1
0.1
-2
-3
0.05
-4
-5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 60 km/h
4 0.25
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
2
0.2
0
0.15
-2
0.1
-4
0.05
-6
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.28 Influence of wheel stiffness on AA responses under shallow bump
101
sharp bump
L 12.967 0.0211 12.015 0.0387 10.805 0.0099
2
BA (m/s ) N 14.609 0.0248 14.569 0.0529 13.588 0.0113
H 15.629 0.0255 17.177 0.0644 16.922 0.0167
L 5.570 0.0083 7.127 0.0104 7.826 0.0075
2
AA (rad/s ) N 6.526 0.0094 8.639 0.0114 9.296 0.0088
H 7.349 0.0101 10.202 0.0123 10.703 0.0098
L:490400 N/m; N: : 613000 N/m; H: 735600 N/m
3.8 Summary
In this chapter, a mathematical model of the passive suspension system of the tracked vehicle is
discussed and its differential equations are derived. A Simulink model is constructed based on the
equations of motion of the suspension system to evaluate the suspension performance. In order to
assess the suspension performance, various bump road profiles are used as input excitations. The
model characteristics are validated using published experimental data for the same tracked vehicle.
The performance of the suspension with different design parameters under different bump road
conditions is investigated. The simulation results reveal that the locations and number of dampers
have a significant effect on the suspension performance over bump terrains. The best suspension
configurations that offer optimal ride are to use two or three dampers in the suspension. The best
locations of these dampers are at the first, second and last wheel stations. In addition, the passive
damper in the suspension has fixed characteristics which limit the suspension performance over
rough roads. In addition, the suspension performance is reduced when the damping is increased.
Furthermore, the suspension with low stiffness offers lower acceleration responses than the stiffer
suspension. Also, the wheel stiffness affects the suspension dynamic behaviour. As the wheel
stiffness is increased, the suspension ride performance is reduced.
102
Speed 10 km/h
1.5 0.2
Low: 490400 N/m Low: 490400 N/m
0.18
High: 735600 N/m High: 735600 N/m
1
0.16
0.14
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.12
0 0.1
0.08
-0.5
0.06
0.04
-1
0.02
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
8 0.35
Low: 490400 N/m Low: 490400 N/m
6 High: 735600 N/m High: 735600 N/m
0.3
4
0.25
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.2
0
-2 0.15
-4
0.1
-6
0.05
-8
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
10 0.25
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
5 0.2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 0.15
-5 0.1
-10 0.05
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.29 Influence of wheel stiffness on BA responses under medium bump
103
Speed 10 km/h
1.5 0.2
Low: 490400 N/m Low: 490400 N/m
0.18
High: 735600 N/m High: 735600 N/m
1
0.16
0.14
0.5

0.12
0 0.1
0.08
-0.5
0.06
0.04
-1
0.02
-1.5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
4 0.2
Low: 490400 N/m Low: 490400 N/m
0.18
3 High: 735600 N/m High: 735600 N/m
0.16
2
0.14
1
0.12
0 0.1
0.08
-1
0.06
-2
0.04
-3
0.02
-4 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
5 0.16
Low: 490400 N/m Low: 490400 N/m
4
High: 735600 N/m 0.14 High: 735600 N/m
3
0.12
2
0.1
1
0 0.08
-1
0.06
-2
0.04
-3
0.02
-4
-5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.30 Influence of wheel stiffness on AA responses under medium bump
104
Speed 10 km/h
8 0.7
Low: 490400 N/m Low: 490400 N/m
6 High: 735600 N/m High: 735600 N/m
0.6
4
0.5
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.4
0
-2 0.3
-4
0.2
-6
0.1
-8
-10 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 4.5
Low: 490400 N/m Low: 490400 N/m
8 High: 735600 N/m 4 High: 735600 N/m
6 3.5
4 3
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2 2.5
0 2
-2 1.5
-4 1
-6 0.5
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
12 0.7
Low: 490400 N/m Low: 490400 N/m
10
High: 735600 N/m High: 735600 N/m
0.6
8
6 0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
4
0.4
0.3
0
-2 0.2
-4
0.1
-6
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.31 Influence of wheel stiffness on BA responses under sharp bump
105
Speed 10 km/h
4 0.18
Low: 490400 N/m Low: 490400 N/m
3 High: 735600 N/m 0.16 High: 735600 N/m
0.14
2
0.12

1
0.1
0
0.08
-1
0.06
-2
0.04
-3 0.02
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
6 0.7
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
0.6
4
0.5
2
0.4
0.3
-2
0.2
-4
0.1
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
6 0.5
Low: 490400 N/m Low: 490400 N/m
0.45
High: 735600 N/m High: 735600 N/m
4
0.4
0.35
2
0.3
0 0.25
0.2
-2
0.15
0.1
-4
0.05
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎3.32 Influence of wheel stiffness on AA responses under sharp bump
106
PERFORMANCE UNDER PERIODIC TERRAINS
4.1 Introduction
This chapter presents the results obtained from the simulation for the suspension performance
under sinusoidal road excitations. It starts with the road characteristics that are utilized in the
simulation. This is followed by the outcome analyses from the parametric study of the suspension
under sinusoidal excitations. The chapter is concluded with a summary of the main findings. It is
noted that, the simulation procedures are done in the same way as in case of the bump excitation.
4.2 Sinusoidal excitations

The road excitation used in this analysis is the periodic road profile in which the unevenness is
represented by a sine wave. The road profile is characterized by its wavelength, ( = 5-8 m) and
height (h= 0.1-0.2 m) as used in some studies [11, 24, 88]. The vertical excitation for each road
wheel varies according to its position from the centre of gravity of the hull. This excitation depends
on the vehicle speed and the road characteristics (wavelength and height). The equation that
th
governs the i road wheel excitation can be expressed as follows:
; ; i=1,2,...,5 (‎4.1)
λ λ
Figure 4.1shows the three different sinusoidal road profiles used in the simulation. The graph
represents the excitation at the first road wheel of the suspension. It can be seen that the signal
does not start from zero as the phase shift mentioned in equation (4.1).
0.1
Shallow
Medium
0.05 Sharp
Amplitude (m)
-0.05
-0.1
0 1 2 3 4
Time (s)
Figure ‎4.1 Typical sinusoidal road profiles at speed of 10 km/h
Table 4.1 Description and parameters of selected road profiles

Road profile type Height (m) Wavelength (m)
Shallow 5
Medium 0.1 2.5
Sharp 1
107
4.3 Parametric analysis of suspension performance under sinusoidal terrains
The analysis is performed through studying the effect of the variation in the suspension design
parameters on vehicle performance under different sinusoidal excitations. The same suspension
parameters discussed in chapter 3 are used here in this simulation.
4.3.1 Influence of damper locations

In this part, the effect of varying the damper locations on the vehicle suspension performance is
investigated. The suspension performance is evaluated under different sinusoidal road excitations
and vehicle speeds. The MOF values of all suspension configurations listed in Table 3.5 are
calculated under shallow, medium and sharp sinusoidal terrains and the data are summarized in
Table 4.2 to Table 4.5.
Table 4.2 shows the MOF values of the suspension configurations S1 under these excitations. The
results denote that the damper locations have a significant effect on the vehicle performance. Over
the shallow sinusoidal road and vehicle speed of 10 km/h, the suspension configuration S1 has two
locations with the lowest MOF values which are at wheel 2 and wheel 1 respectively. In addition,
when the vehicle traverses the same road with 40 km/h and 60 km/h, the suspension fitted with a
damper at wheel 1 or wheel 4 offers the lowest MOF values. Over the medium sinusoidal profile
and vehicle speed of 10 km/h, the lowest MOF values occur when the suspension is equipped with
dampers at wheel 4 and wheel 1. For the same road, when the vehicle speed is increased to 40
and 60 km/h, the suspension with dampers at wheel 3 or wheel 2 and wheel 3 or wheel 5 has the
lowest MOF values. Over the sharp sinusoidal profile with speeds 10 and 40 km/h, the best
locations are at wheel 3 or wheel 2 and wheel 5 or wheel 2 while at speed 60 km/h, the best
damper locations are at wheel 4 or wheel 1.
Table 4.3 shows the MOF values of the suspension configurations S2 under the same excitations.
The results indicate the significant effect of the damper locations on the vehicle performance. Over
the shallow sinusoidal road profile and a speed of 10 km/h, the suspensions with dampers at
wheels 12 and wheels 15 are the best configurations that have the lowest MOF values. When the
vehicle traverses the same road with speed 40 km/h, the best suspension configurations are at
wheels 45 and wheels 34. At 60 km/h, dampers at wheels 15 and wheels 34 have the lowest MOF
values. Over the medium sinusoidal road profile, the suspension with dampers at wheels 23 and
wheels 45 has the lowest MOF values at 10 km/h while dampers at wheels 13 and wheels 15 have
the best locations at speed 40 km/h. At 60 km/h, dampers at wheels 13 and wheels 35 have the
best locations that have the lowest MOF values. Over the sharp sinusoidal road profile and at
speeds of 10 and 40 km/h, dampers at wheels 34 and wheels 35 are the best locations, while
dampers at wheels 14 and wheels 24 have the best damper locations at speed 60 km/h.
Likewise, the MOF values of all the suspension configurations S3 are shown in Table 4.4. Over the
shallow sinusoidal road profile and a speed of 10 km/h, the suspensions with dampers at wheels
125 and wheels 124 are the best configurations that have the lowest MOF values. When the
vehicle traverses the same road with speed 40 km/h, the best suspension configurations are at
wheels 345 and wheels 125. At 60 km/h, dampers at wheels 345 and wheels 125 have the lowest
MOF values. Over the medium sinusoidal road profile, the suspension with dampers at wheels 134
108
and wheels 123 has the lowest MOF values at 10 km/h while dampers at wheels 125 and wheels
134 have the best locations at speed 40 km/h. At 60 km/h, dampers at wheels 125 and wheels 235
have the best locations that have the lowest MOF values. Over the sharp sinusoidal road profile
and at speeds of 10 km/h, dampers at wheels 345 and wheels 234 are the best locations, while
dampers at wheels 123 and wheels 345 have the best damper locations at speed 40 and 60 km/h.
Table 4.2 MOF values of configurations S1 under different sinusoidal excitations

Suspensio Shallow Medium Sharp
n confg. V1 V2 V3 V1 V2 V3 V1 V2 V3
undamped 0.238 0.243 0.244 0.238 0.241 0.240 0.240 0.231 0.247
Wh1 0.130 0.232 0.242 0.240 0.294 0.260 0.275 0.236 0.213
Wh2 0.129 0.269 0.250 0.264 0.279 0.272 0.262 0.221 0.217
Wh3 0.210 0.261 0.253 0.249 0.246 0.230 0.249 0.239 0.232
Wh4 0.179 0.236 0.230 0.189 0.292 0.276 0.263 0.220 0.202
Wh5 0.169 0.237 0.246 0.253 0.286 0.236 0.300 0.220 0.227

undamped 0.238 0.243 0.244 0.238 0.241 0.240 0.240 0.231 0.247
Wh12 0.116 0.276 0.302 0.295 0.295 0.286 0.282 0.244 0.230
Wh13 0.123 0.305 0.287 0.286 0.258 0.237 0.308 0.240 0.247
Wh14 0.134 0.314 0.272 0.255 0.287 0.271 0.286 0.236 0.222
Wh15 0.119 0.310 0.245 0.290 0.270 0.253 0.311 0.250 0.242
Wh23 0.124 0.282 0.281 0.211 0.317 0.297 0.267 0.250 0.243
Wh24 0.178 0.294 0.281 0.294 0.300 0.286 0.275 0.245 0.225
Wh25 0.152 0.306 0.286 0.274 0.285 0.246 0.301 0.245 0.242
Wh34 0.175 0.252 0.251 0.250 0.275 0.268 0.264 0.220 0.229
Wh35 0.165 0.257 0.268 0.282 0.272 0.239 0.266 0.225 0.222
Wh45 0.152 0.185 0.272 0.248 0.302 0.262 0.324 0.242 0.235

undamped 0.238 0.243 0.244 0.238 0.241 0.240 0.240 0.231 0.247
Wh123 0.113 0.291 0.315 0.223 0.296 0.276 0.298 0.222 0.227
Wh124 0.099 0.302 0.316 0.300 0.304 0.299 0.298 0.234 0.233
Wh125 0.082 0.233 0.302 0.289 0.273 0.246 0.315 0.263 0.286
Wh134 0.118 0.326 0.315 0.221 0.276 0.264 0.304 0.265 0.250
Wh135 0.112 0.323 0.319 0.301 0.279 0.270 0.313 0.250 0.263
Wh145 0.097 0.301 0.322 0.252 0.277 0.264 0.321 0.249 0.263
Wh234 0.152 0.301 0.299 0.284 0.307 0.286 0.275 0.245 0.241
Wh235 0.150 0.319 0.314 0.221 0.288 0.253 0.296 0.239 0.232
Wh245 0.131 0.281 0.312 0.291 0.310 0.280 0.303 0.253 0.254
Wh345 0.153 0.202 0.276 0.291 0.280 0.254 0.273 0.226 0.230
V1= 10 km/h; V2= 40 km/h; V3= 60 km/h
109
Table 4.5 lists the MOF values of the suspension configurations S4 and S5. Over the shallow
sinusoidal road profile and at 10 km/h, the suspension S4 has two locations with the lowest MOF
which are at wheels 1235 and wheels 1245 while dampers at wheels 2345 and wheels 1345 have
the best locations at speed 40 km/h. At high speed, dampers at wheels 2345,1235 and 1245 have
the lowest MOF values. Over the medium sinusoidal road profile, wheels 1245 and wheels 1345
are the best damper locations at 10 km/h while wheels 1235 and wheels 1234 are the best damper
locations at 40 km/h. At speed of 60 km/h, wheels 2345 and wheels 1234 are the best damper
locations. Over the sharp sinusoidal road profile and at speeds of 10 km/h, dampers at wheels
2345 and wheels 1234 are the best locations, while dampers at wheels 2345 and wheels 1234
have the best damper locations at speed 40 km/h and 60 km/h.
It can be seen from the results that the damper locations affect the vehicle performance and the
best damper locations over specific sinusoidal road profile are changed when the vehicle speed is
varied. By looking to the best configurations with the lowest MOF, it is found that specific wheels
are repeated in each configuration. In instance, wheel number 1 is repeated five times in the best
damper locations in the suspension configuration S1. In addition, wheels 1 and 5 are repeated 7
and 8 times in the best locations of the suspension configuration S2. Thus, it is found that the
overall best locations of the suspension configurations are at wheel 1, wheels 15, wheels 125 and
wheels 1245.
Table ‎4.5 MOF values of configurations S4 and S5 under different sinusoidal excitations
undamped 0.238 0.243 0.244 0.238 0.241 0.240 0.240 0.231 0.247
Wh1234 0.098 0.319 0.327 0.274 0.299 0.278 0.301 0.248 0.252
Wh1235 0.081 0.314 0.338 0.248 0.292 0.279 0.326 0.259 0.281
Wh1245 0.086 0.315 0.338 0.226 0.327 0.330 0.325 0.286 0.288
Wh1345 0.093 0.300 0.331 0.242 0.305 0.289 0.321 0.252 0.279
Wh2345 0.126 0.287 0.293 0.314 0.300 0.267 0.287 0.228 0.223
All wheels 0.081 0.320 0.347 0.229 0.338 0.339 0.341 0.308 0.295
V1= 10 km/h; V2= 40 km/h; V3= 60 km/h
4.3.2 Influence of number of dampers

This section presents the effect of the number of dampers in the suspension on the vehicle
performance under sinusoidal road profiles. The suspension performance is presented in the same
way as bump excitation analysis. The time and frequency histories of the BA and AA responses of
the best suspension locations that have been selected in section 3.7.2 are presented and
compared with the undamped response under the same excitations and speeds.
1- Shallow sinusoidal responses
The system responses of the suspension system with the best configurations under the shallow
sinusoidal excitation are shown in Figure 4.2(a, b). The time and frequency domain responses of
BA and AA responses of all suspension configurations are compared with those for the undamped
suspension at different speeds of 10, 40 and 60 km/h. The results from Figure 4.2(a) show that the
damped suspension dissipates the energy caused by the shallow sinusoidal excitation and
110
improves the suspension performance at all speeds. Figure 4.2(b) shows the absolute values of the
FFT responses of the BA responses under the shallow excitations. At the low speed of 10 km/h,
the excitation frequency is 0.55 Hz. In the undamped case, this excites the pitch and bounce peak
response amplitudes at frequencies of 0.9 Hz and 1.5 Hz, respectively, as well as the wheel
bounce amplitude at frequency of 12.5 Hz. For the damped suspensions, the pitch and bounce
peak amplitudes are suppressed. At the higher speeds of 40 km/h and 60 km/h, only the bounce
response is excited even though the excitation frequencies are higher. However, the pitch
response is not excited because the excitation frequencies are 2.22 Hz and 3.33 Hz, respectively.
The PTP values of the BA of all suspension configurations are listed in Table 4.6. The data indicate
that the amount of the damping force and the vehicle speed have a significant effect on the
suspension performance. For instance, at 10 km/h the suspension configurations S2-15 and S4-
1245l have the highest reduction in the PTP values, followed by S5-all. However, suspension
configurations S1-1 and S3-125 increase the PTP values and reduce the suspension performance.
When the vehicle negotiates the same road at 40 km/h, increasing the number of dampers up to
three improves the suspension performance. However, the suspension configurations S4-1245 and
S5-all did not offer significant effect on the suspension behaviour. At 60 km/h, the road-induced
vibration transmitted to the vehicle body is significant. Therefore, the suspension should isolate the
vehicle body from these vibrations. It is found that increasing the number of dampers not always
improves the suspension behaviour. This agrees with the results obtained from the analysis of the
suspension under bump terrains in section3.7.3. The PTP values at speed 60 km/h show that the
suspension configurations S1-1 and S2-15 give the highest reduction in the suspension response.
The RMS values of the BA responses are listed in Table 4.6. The results from the table show that
the configurations S4-1245 and S5-all have the lowest RMS values under the shallow sinusoidal
road profile at 10 km/h. However, over the medium and sharp sinusoidal profiles, the suspension
configurations S1-1 and S3-125 and S2-15 and S3-125 are superior in enhancing the vehicle
performance.
the same excitations. The results shown in Figure 4.3(a) demonstrate that the damped suspension
configurations have a better performance than the undamped suspension S0 for at low speed only.
However, at medium and high speeds, the damped suspension increases the AA responses
compared with the undamped suspension.
Figure 4.3(b) shows the absolute values of the FFT responses of the AA responses under the
shallow excitations. At the low speed of 10 km/h, the undamped response shows the pitch peak
amplitude at frequencies of 0.9 Hz as well as the wheel bounce amplitude at frequency of 12.5 Hz.
The damped suspension shows peak amplitudes at the excitation frequency of 0.55 Hz, but
suppressed amplitudes at the pitch frequency. For speeds of 40 km/h and 60 km/h, the pitch
response is slightly excited but the bounce response is not excited because the excitation
frequencies are 2.22 Hz and 3.33 Hz. The figure also shows that the damped suspensions have
reduced the AA resonant peaks at the pitch frequency of 0.9 Hz at all speeds. Table 4.7 shows the
111
PTP values of the AA responses under the same excitations and speeds. The table indicates that
the suspensions with up to three dampers have better responses than the undamped suspension.
The suspension configurations S1-1, S2-15 and S3-125 have the biggest reductions in the PTP
values. However, the PTP values of the AA responses at 40 and 60 km/h are increased when the
number of dampers is increased. In addition the suspensions with up to three dampers still have
the lowest increase in the AA responses. All the suspension configurations have reduced the RMS
values of the AA responses at low speed. However, all damped suspension configurations increase
the AA responses significantly at 40 km/h and 60 km/h.
112
Speed 10 km/h
10 1.8
Undamped Undamped
At wh1 At wh1
At wh15 1.6 At wh15
7.5 At wh125 At wh125
At wh1245 At wh1245
1.4
Bounce Acc (m/s2) 5 1.2
Bounce Acc (m/s2)

1
2.5
0.8
0 0.6
0.4
-2.5
0.2
-5 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
15 10
Undamped Undamped
At wh1 At wh1
9
At wh15 At wh15
10 At wh125 At wh125
At wh1245 8 At wh1245
7
5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

6
0 5
-5
3
2
-10
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
25 20
Undamped Undamped
At wh1 At wh1
20 18
At wh15 At wh15
At wh125 At wh125
15 At wh1245 16 At wh1245
10 14
Bounce Acc (m/s2)
Bounce Acc (m/s2)
5 12
0 10
-5 8
-10 6
-15 4
-20 2
-25 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.2 Bounce acceleration responses under shallow sinusoidal excitation
113
Speed 10 km/h
15 4
Undamped Undamped
At wh1 At wh1
At wh15 3.5 At wh15
At wh125 At wh125
10 At wh1245 At wh1245
At all wheels 3 At all wheels

2.5
5
0
1.5
1
-5
0.5
-10 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
15 10
Undamped Undamped
At wh1 At wh1
9
At wh15 At wh15
At wh1245 8 At wh1245
7
5
0 5
-5
3
2
-10
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
15 12
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
10 At wh125 10 At wh125
At wh1245 At wh1245
5 8
0 6
-5 4
-10 2
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.3 Angular acceleration responses under shallow sinusoidal excitation
114
Table 4.6 PTP and RMS values of BA under shallow sinusoidal excitation
-2 -2
PTP (ms ) RMS (ms )
Description
V=10 PTP% V=40 PTP% V=60 PTP% V=10 RMS% V=40 RMS% V=60 RMS%
S0 5.84 0 23.13 0 15.84 0 0.0275 0 0.1157 0 0.0772 0
S1-1 12.81 -119.4 13.99 39.5 15.52 2.0 0.0275 0.04 0.0561 51.6 0.0747 3.2
S2-15 2.08 64.4 13.47 41.7 12.20 23.0 0.0101 63.12 0.0828 28.5 0.0663 14.1
S3-125 10.10 -72.9 12.01 48.0 16.70 -5.4 0.0184 32.93 0.0499 56.9 0.0999 -29.5
S4-1245 2.10 64.1 18.34 20.7 29.13 -83.8 0.0080 70.76 0.1242 -7.3 0.1977 -156.2
S5-all 3.17 45.8 21.36 7.6 43.10 -172.1 0.0086 68.64 0.1498 -29.4 0.3024 -292.0
Table 4.7 PTP and RMS values of AA under shallow sinusoidal excitation
-2 -2
Description
S0 16.42 0 12.54 0 11.70 0 0.0715 0 0.0578 0 0.0564 0
S1-1 10.26 37.5 13.16 -4.9 15.64 -33.7 0.0335 53.1 0.0721 -24.9 0.0877 -55.5
S2-15 13.74 16.3 18.63 -48.5 20.46 -74.8 0.0262 63.3 0.1100 -90.4 0.1341 -137.8
S3-125 15.51 5.5 21.90 -74.6 24.36 -108.2 0.0257 64.0 0.1319 -128.4 0.1665 -195.3
S4-1245 17.09 -4.1 23.79 -89.6 26.18 -123.8 0.0255 64.3 0.1406 -143.3 0.1779 -215.4
S5-all 17.05 -3.9 23.95 -90.9 26.52 -126.6 0.0255 64.3 0.1443 -149.7 0.1841 -226.4
115
2- Medium sinusoidal responses
Over the medium sinusoidal road profile, the excitation frequency is increased compared to the
shallow sinusoidal road profile. The increase in the road frequency introduces high vibration levels
to the suspension system. Figure 4.4(a, b) shows the system responses of the BA of all suspension
configurations. From Figure 4.4(a), the results indicate that all the damped suspension
configurations have better acceleration responses than the undamped suspension at low speed
only. However, at medium and high speeds the damped suspension increases the response.
Figure 4.4(b) shows the frequency domain responses of all suspension configurations under the
same excitations. At speed of 10 km/h, the undamped suspension shows bounce peak amplitudes
at the excitation frequency of 1.11 Hz and at the bounce frequency of 1.5 Hz. The damped
suspension also shows a reduced peak amplitude at excitation frequency of 1.11 Hz but no peak
response at the bounce frequency. At higher speeds of 40 km/h and 60 km/h, the bounce response
is very slightly excited; the peak amplitudes are due to the excitation frequencies of 4.44 Hz and
6.66 Hz respectively.
The PTP values of the BA for all suspension configurations at various speeds are summarized in
Table 4.8. At 10 km/h, all the damped suspension configurations have better reduction in the PTP
values of BA than the undamped suspension. Also, the percent reduction is increased when the
suspension have more than 3 dampers. This appears from the table as the suspension
configurations S5-all and S4-1245 have the highest reduction in the PTP values. It is noticed that
when the speed is increased to 40 km/h and 60 km/h, the damped suspension configurations have
higher PTP values than the undamped suspension. The RMS values of the BA responses are
listed in Table 4.8. The table shows that at 40 km/h and 60 km/h, the damped suspension has
poorer behaviour than the undamped suspension. In addition, when the number of dampers is
increased the suspension responses are increased.
Similarly, the AA responses under the medium excitation are shown in Figure 4.5(a, b). It is seen
that all the damped suspension configurations have a significant effect on the suspension
performance. The damped suspension improves the AA responses at low speed. Figure 4.5(b)
illustrates the frequency domain of the AA responses under the same excitation. At the low speed
of 10 km/h, the undamped response shows peak amplitudes at the pitch frequency of 0.9 Hz and
excitation frequency of 1.11 Hz as well as the wheel bounce peak amplitude at frequency of 12.5
Hz. The damped suspension shows peak amplitudes only at the excitation frequency of 1.11 Hz
and wheel bouncy frequency of 12.5 Hz. For speeds of 40 km/h and 60 km/h, peak amplitudes
occur only at the excitation frequencies of 4.44 Hz and 6.66 Hz. There are no peak amplitudes at
the pitch resonance frequency of 0.9 Hz because the excitation frequencies are higher. Although
the damped suspension has reduced the AA peak at the wheel resonant frequency of 12.5 Hz at all
speeds, the undamped suspension has a better performance in the frequency range lower than 12
Hz.
116
The PTP values of the AA are listed in Table 4.9. At 10 km/h, the damped suspension performance
has improved. In addition, the suspension configurations S3-125 and S1-1 offer the highest
reduction in the PTP values of AA. This is followed by the suspension configurations S5-all, S4-
1245 and S2-15. For speeds 40 and 60 km/h, the undamped suspension has the lowest peak
values. Table 4.9 summarizes the RMS values of the AA responses at all speeds. It is seen that
the damped suspension configurations have better performance than the undamped suspension at
low speed. However, the undamped suspension is superior in reducing the vibration at medium
and high speeds.
117
8 4
Undamped Undamped
At wh1 At wh1
6 At wh15 3.5 At wh15
At wh125 At wh125
At wh1245 At wh1245
4 At all wheels 3 At all wheels
2 2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0 2
-2 1.5
-4 1
-6 0.5
-8 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 40 km/h
30 25
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
At wh1245 20 At wh1245
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
15
10
-10
5
-20
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
30 25
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
At wh1245 20 At wh1245
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
15
10
-10
5
-20
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.4 Bounce acceleration responses under medium sinusoidal excitation
118
Speed 10 km/h
3 1.4
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
1.2
2 At wh125 At wh125
At wh1245 At wh1245
1
1

0.8
0.6
-1
0.4
-2
0.2
-3 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 40 km/h
6 4.5
Undamped Undamped
At wh1 At wh1
At wh15 4 At wh15
4 At wh125 At wh125
At wh1245 At wh1245
3.5
2 3
2.5
0
2
-2 1.5
1
-4
0.5
-6 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
8 5
Undamped Undamped
At wh1 At wh1
4.5
6 At wh15 At wh15
At wh125 At wh125
At wh1245 4 At wh1245
3.5
2
3
0 2.5
2
-2
1.5
-4
1
-6
0.5
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.5 Angular acceleration responses under medium sinusoidal excitation
119
Table ‎4.8 PTP and RMS values of BA under medium sinusoidal excitation
-2 -2
S0 12.57 0 8.45 0 9.86 0 0.0630 0 0.0402 0 0.0473 0
S1-1 7.91 37.0 22.64 -167.9 25.22 -155.7 0.0456 27.5 0.1477 -267.2 0.1652 -249.1
S2-15 10.03 20.2 42.23 -399.7 44.89 -355.1 0.0556 11.7 0.2769 -588.2 0.2973 -528.0
S3-125 12.04 4.2 53.79 -536.4 56.42 -472.1 0.0622 1.2 0.3484 -766.0 0.3581 -656.5
S4-1245 7.26 42.3 51.43 -508.6 54.62 -453.8 0.0391 37.9 0.3466 -761.5 0.3542 -648.1
S5-all 4.46 64.5 30.17 -257.0 31.48 -219.2 0.0217 65.6 0.2098 -421.5 0.2132 -350.3
Table ‎4.9 PTP and RMS values of AA under medium sinusoidal excitation
-2 -2
S0 5.67 0 2.44 0 2.86 0 0.0240 0 0.0118 0 0.0142 0
S1-1 3.61 36.3 10.78 -341.2 12.23 -326.7 0.0134 43.9 0.0612 -417.3 0.0690 -387.3
S2-15 4.20 26.0 5.01 -105.0 5.45 -90.2 0.0207 13.7 0.0250 -111.8 0.0269 -90.3
S3-125 3.38 40.4 3.31 -35.4 4.02 -40.2 0.0182 23.9 0.0176 -48.4 0.0216 -52.9
S4-1245 4.06 28.5 5.97 -144.4 6.98 -143.5 0.0166 30.7 0.0421 -256.0 0.0489 -245.8
S5-all 3.81 33.0 5.57 -127.8 6.72 -134.5 0.0157 34.6 0.0392 -231.8 0.0473 -234.0
120
3- Sharp sinusoidal responses
In this section, the suspension performance is investigated under severe sinusoidal excitation with
0.5 m wavelength (too short for full-size tracked vehicles). The BA and AA of the undamped and
damped suspension configurations are shown in Figure 4.6(a, b) and Figure 4.7(a, b). The BA
responses shown in (a) indicate that under the sharp terrains the suspension performance is highly
influenced by the number of dampers fitted to the suspension. The suspension responses are
significantly increased when the suspension is fitted with any dampers. Figure 4.6(b) shows the
FFT responses of the suspension systems under the same excitations. At speed of 10 km/h, the
undamped suspension shows peak amplitudes at the bounce frequency of 1.5 Hz, excitation
frequency of 2.77 Hz, and wheel frequency of 12.5 Hz. For speeds of 40 km/h and 60 km/h, the
bounce response is only very slightly excited because the excitation frequencies are 11.11 Hz and
16.16 Hz, respectively. But at these high speeds clear peaks are seen at the excitation and wheel
bounce frequencies. Also, the figure shows that that at 10 km/h, the damped suspension reduces
the BA resonant peaks while the response is increased after the resonant frequency (1.5 Hz).
When the vehicle traverses over the same sharp sinusoidal road at the medium speed of 40 km/h,
the S2-15 and S5-all suspension configurations have reduced the BA peaks while the other
configurations increase the BA peaks as shown in Figure 4.6(a). However, it is clear from
Figure 4.6(b) that this increase in the BA peaks is due to the wheel bounce of frequency 12.6 Hz.
Similarly, when the vehicle traverses the sharp sinusoidal at the high speed of 60 km/h, the
damped suspension responses are increased compared with the undamped one
Table 4.10 lists the PTP values of the BA responses of the suspension configurations under sharp
sinusoidal excitation. It can be seen that the BA responses are significantly increased when the
suspension is equipped with any number of dampers. This shows that the increase in the damping
forces is not always favourable to vibration attenuation especially over rough sinusoidal roads.
The time history of the AA of all suspension configurations under the same sharp sinusoidal road
excitation is illustrated in Figure 4.7(a). The responses of the damped suspension configurations
have higher responses than the undamped suspension at all speeds. The frequency domain
responses shown in Figure 4.7(b) clearly shows peak amplitude for the undamped and damped
suspensions due to the excitation frequency of 2.77 Hz. In the undamped case, a very small peak
is also seen at the pitch frequency of 0.9 Hz. At higher speeds of 40 km/h and 60 km/h, the pitch
response is not excited because the excitation frequencies are 11.11 Hz and 16.16 Hz. But at
these high speeds clear peaks are seen at the excitation and wheel bounce frequencies. The figure
also indicates that the wheel resonant peaks are reduced at 10 km/h and 40 km/h. However, at 60
km/h the damped suspension responses have significant amplitudes at the wheel resonant
frequency (12.7 Hz). The results listed in Table 4.11 show the RMS values of the AA under
different speeds. The data show that all the damped suspension configurations have increased the
RMS values of the AA responses at all speeds.
121
20 14
Undamped Undamped
At wh1 At wh1
15 At wh15 At wh15
12
At wh125 At wh125
At wh1245 At wh1245
10
5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

8
6
-5
4
-10
2
-15
-20 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 40 km/h
25 16
Undamped Undamped
At wh1 At wh1
20
At wh15 14 At wh15
At wh125 At wh125
15 At wh1245 At wh1245
10
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 8
-5
6
-10
4
-15
2
-20
-25 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
30 25
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
At wh1245 20 At wh1245
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
15
10
-10
5
-20
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure ‎4.6 Bounce acceleration responses under sharp sinusoidal excitation
122
Speed 10 km/h
8 6
Undamped Undamped
At wh1 At wh1
6 At wh15 At wh15
At wh125 5 At wh125
At wh1245 At wh1245
4

2
0 3
-2
2
-4
1
-6
-8 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 40 km/h
15 8
Undamped Undamped
At wh1 At wh1
At wh15 7 At wh15
At wh1245 At wh1245
5
Angular Acc (rad/s2) 5
0 4
3
-5
-10
1
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
15 12
Undamped Undamped
At wh1 At wh1
At wh15 At wh15
10 At wh125 10 At wh125
At wh1245 At wh1245
5 8
0 6
-5 4
-10 2
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure 4.7 Angular acceleration responses under sharp sinusoidal excitation
123
Table 4.10 PTP and RMS values BA under sharp sinusoidal excitation
-2 -2
PTP (ms ) RMS (ms )
Description
S0 9.45 0 30.82 0 11.84 0 0.0439 0 0.1530 0 0.0582 0
S1-1 26.07 -175.9 37.30 -21.0 30.92 -161.1 0.1607 -266.2 0.1928 -26.0 0.1871 -221.4
S2-15 21.81 -130.8 26.79 13.1 25.38 -114.3 0.1469 -234.7 0.1778 -16.2 0.1684 -189.2
S3-125 34.16 -261.5 46.75 -51.7 49.29 -316.2 0.2133 -385.9 0.2399 -56.8 0.3133 -438.0
S4-1245 32.55 -244.5 46.61 -51.2 54.43 -359.7 0.2215 -404.8 0.2402 -56.9 0.3400 -483.9
S5-all 17.79 -88.3 27.36 11.2 25.37 -114.2 0.1249 -184.6 0.1795 -17.3 0.1648 -183.1
Table 4.11 PTP and RMS values of AA under sharp sinusoidal excitation
-2 -2
Description
S0 3.64 0 13.57 0 4.46 0 0.0168 0 0.0671 0 0.0219 0
S1-1 8.28 -127.4 14.81 -9.1 14.92 -234.4 0.0532 -216.9 0.0825 -23.1 0.0948 -333.3
S2-15 13.89 -281.3 21.91 -61.5 25.40 -469.5 0.0876 -421.5 0.1159 -72.9 0.1635 -647.2
S3-125 8.43 -131.4 16.86 -24.2 18.39 -312.3 0.0525 -212.4 0.0938 -39.9 0.1183 -440.8
S4-1245 7.29 -100.2 10.58 22.0 10.12 -126.9 0.0463 -175.5 0.0738 -10.0 0.0705 -222.4
S5-all 6.78 -86.3 11.14 17.9 10.71 -140.2 0.0457 -172.0 0.0769 -14.6 0.0734 -235.7
124
4.3.3 Influence of damping coefficient
In this section, the effect of varying the damping coefficient of the passive damper on the vehicle
performance under sinusoidal road profiles is studied. The results from the simulation are
presented in the same manner like the bump excitation. The vehicle configurations and parameters
are fixed the type of excitation which is the different sinusoidal excitation mentioned in section 4.2.
The influence of the variations in the suspension damping is shown in Figure 4.8(a, b) and
Figure 4.9(a, b). The BA responses in Figure 4.8(a) reveals that when the damping is increased the
BA responses are increased during all vehicle speeds. Moreover, the BA spectra presented in
Figure 4.8(b) show peak amplitudes at excitation frequencies 0.5 Hz, 2.22 Hz and 3.33 Hz for the
suspension responses at speeds of 10, 40 and 60 km/h, respectively, as well as at the wheel
bounce frequencies. Also, the figure demonstrates that the increase in the suspension damping
offers poorer isolation at the neighbourhood of bounce resonant frequency. This can be noticed
from Table 4.12 which shows the increase in the RMS values of the BA at all speeds.
the same sinusoidal excitation and speeds. The AA responses shown in Figure 4.9(a ) show that
as the damping increased, the suspension performance is reduced at all speeds. This appears in
the increase in the PTP values of AA shown in Table 4.12. The FFT plot shows the peak
amplitudes at excitation frequency of 0.5 Hz, pitch frequency of 0.9 Hz and at wheel bounce
frequency of 12.5 Hz at speed of 10 km/h. For speeds of 40 and 60 km/h, the peak amplitudes are
due to the excitation frequencies of 2.22 Hz and 3.33 Hz, and the wheel bounce frequency of 12.5
Hz. Also, the AA responses show the same trend as the BA responses where the increase in the
damping offers poor isolation and an increase in the RMS values as shown in Table 4.12.
Figure 4.10(a, b) show the BA of the system responses in time and frequency domains under
medium sinusoidal excitation. The time history of the BA shown in Figure 4.10(a) demonstrates that
increasing the suspension damping degrades the suspension performance for all vehicle speeds.
In addition, the BA spectra shown in Figure 4.10(b) indicate peak amplitudes at the bounce
frequency of 1.11 Hz in addition to wheel bounce at frequency of 12.5 Hz. Also, the figure shows
an increase in the suspension responses. The PTP values of the BA responses are significantly
increased compared with the case of the shallow sinusoidal excitation as shown in Table 4.13.
Similarly, the AA responses under the same excitation are shown in Figure 4.11(a, b). The AA time
responses indicate that the suspension response is increased by increasing the damping force.
Furthermore, Figure 4.11(b) show the FFT responses which have peak amplitudes at pitch
frequency of 0.9 Hz and wheel bounce amplitude at frequency of 12.5 Hz at a speed of 10 km/h.
For speeds of 40 and 60 km/h, the suspension has peak amplitudes at excitation frequencies of
4.44 Hz and 6.66 Hz. Also, the AA responses demonstrate no significant improvement in the
suspension performance when the damping is increased as shown in Table 4.13.
125
This section presents the results obtained for the simulation of the suspension with different
damping values under sharp sinusoidal excitation. The BA and AA responses of the suspension
system under this sharp excitation are shown in Figure 4.12(a, b) and Figure 4.13(a, b). The BA
responses shown in Figure 4.12(a) indicate that the increase in the suspension gives a significant
increase in the suspension responses. The FFT responses at speeds of 10 km/h, 40 km/h and 60
km/h show peak amplitudes at the excitation frequencies of 2.77 Hz, 11.11 Hz and 16.16 Hz,
respectively, and at the wheel bounce frequency of 12.5 Hz. The BA spectra indicate an increase in
the RMS values of the BA responses when the damping is increased. The PTP values of the BA
listed in Table 4.14 confirm that the PTP values are significantly increased compared with the
shallow and medium sinusoidal excitation when the damping is increased at all speeds.
Similarly, the AA responses under the same excitation are shown in Figure 4.13(a, b). The AA time
responses indicate that the suspension response is increased by increasing the damping force.
Furthermore, Figure 4.13(b) show the FFT responses of the suspension under the same excitation.
The peak amplitudes seen at speeds of 10 km/h, 40 km/h and 60 km/h are due to the
corresponding excitation frequencies of 2.77 Hz, 11.11 Hz and 16.16 Hz, respectively, as well as at
the wheel bounce frequency of 12.5 Hz. Also, the AA responses demonstrate no significant
improvement in the suspension performance when the damping is increased as shown in
Table 4.14.
126
Speed 10 km/h
8 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.2
6
1
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
0
0.4
-2
0.2
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
8 3.5
Low: 18016 Ns/m Low: 18016 Ns/m
6 3
4 2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2 2
0 1.5
-2 1
-4 0.5
-6 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
15 8
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 7 High: 27024 Ns/m
10
5
5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 4
3
-5
-10
1
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.8 Influence of varying damping coefficient on BA responses under shallow sinusoidal road
profile
127
Speed 10 km/h
14 1.5
Low: 18016 Ns/m Low: 18016 Ns/m
12 High: 27024 Ns/m High: 27024 Ns/m
10
8 1

6
2 0.5
-2
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
15 10
Low: 18016 Ns/m Low: 18016 Ns/m
9
10
8
7
5
0 5
-5
3
2
-10
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
15 14
Low: 18016 Ns/m Low: 18016 Ns/m
12
10
10
5
-5
4
-10
2
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.9 Influence of varying damping coefficient on AA responses under shallow sinusoidal road
profile
128
Speed 10 km/h
8 4
Low: 18016 Ns/m Low: 18016 Ns/m
6 High: 27024 Ns/m 3.5 High: 27024 Ns/m
4 3
Bounce Acc (m/s2) 2 2.5
Bounce Acc (m/s2)

0 2
-2 1.5
-4 1
-6 0.5
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
40 30
Low: 18016 Ns/m Low: 18016 Ns/m
30 High: 27024 Ns/m High: 27024 Ns/m
25
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 15
-10
10
-20
5
-30
-40 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
30 30
Low: 18016 Ns/m Low: 18016 Ns/m
20
25
10
20
Bounce Acc (m/s2)
Bounce Acc (m/s2)
15
-10
10
-20
5
-30
-40 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.10 Influence of varying damping coefficient on BA responses under medium sinusoidal
road profile
129
Speed 10 km/h
2 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.5 High: 27024 Ns/m High: 27024 Ns/m
1.2
1
1

0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
1.5 1
Low: 18016 Ns/m Low: 18016 Ns/m
0.9
1
0.8
0.7
0.5
0.6
0 0.5
0.4
-0.5
0.3
0.2
-1
0.1
-1.5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
2 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.5 High: 27024 Ns/m High: 27024 Ns/m
1.2
1
1
0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.11 Influence of varying damping coefficient on AA responses under medium sinusoidal
road profile
130
Speed 10 km/h
20 15
Low: 18016 Ns/m Low: 18016 Ns/m
15 High: 27024 Ns/m High: 27024 Ns/m
10
10
Bounce Acc (m/s2) 5
Bounce Acc (m/s2)

0
-5
5
-10
-15
-20 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
25 14
Low: 18016 Ns/m Low: 18016 Ns/m
20
12
15
10 10
Bounce Acc (m/s2)
5 Bounce Acc (m/s2)

8
6
-5
-10 4
-15
2
-20
-25 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
25 25
Low: 18016 Ns/m Low: 18016 Ns/m
20
15 20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
5 15
-5 10
-10
-15 5
-20
-25 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure ‎4.12 Influence of varying damping coefficient on BA responses under sharp sinusoidal road
profile
131
Speed 10 km/h
8 4
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 3.5 High: 27024 Ns/m
6
3
4

2.5
2
0
1.5
-2
1
-4
0.5
-6 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 7
Low: 18016 Ns/m Low: 18016 Ns/m
8
6
6
4 5
2
4
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
10 9
Low: 18016 Ns/m Low: 18016 Ns/m
8 8
6
7
4
6
2
5
0
4
-2
3
-4
2
-6
-8 1
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure 4.13 Influence of varying damping coefficient on AA responses under sharp sinusoidal road
profile
132
Table ‎4.12 Effect of damping on PTP and RMS values under shallow sinusoidal excitation
Damping
Response V=10 V=40 V=60
(Ns/m)
L 8.836 0.0201 11.483 0.0517 13.900 0.0785
2
BA (m/s ) N 9.826 0.0203 12.222 0.0534 16.925 0.1000
H 10.614 0.0205 12.873 0.0575 20.486 0.1243
L 14.662 0.0260 19.797 0.1109 21.820 0.1386
AA
2 N 15.467 0.0258 21.922 0.1320 24.302 0.1649
(rad/s )
H 16.336 0.0258 23.850 0.1529 27.610 0.1892
L:18016; N: : 22520; H: 27024
Table ‎4.13 Effect of damping on PTP and RMS values under medium sinusoidal excitation
Damping Vehicle speed (km/h)
Response (Ns/m) V=10 V=40 V=60
L 12.582 0.0635 47.261 0.3019 53.448 0.3403
2
BA (m/s ) N 13.861 0.0639 55.108 0.3533 57.507 0.3633
H 14.910 0.0640 61.808 0.3969 60.312 0.3781
L 3.658 0.0199 2.116 0.0104 3.337 0.0167
2
AA (rad/s ) N 3.466 0.0187 2.495 0.0125 3.578 0.0181
H 3.306 0.0177 2.924 0.0151 3.767 0.0192
L:18016; N: : 22520; H: 27024
Table ‎4.14 Effect of damping on PTP and RMS values under sharp sinusoidal excitation
Damping Vehicle speed (km/h)
Response (Ns/m) V=10 V=40 V=60
L 26.812 0.1630 43.381 0.2229 47.810 0.3020
2
BA (m/s ) N 32.654 0.2051 43.568 0.2240 48.448 0.3067
H 38.540 0.2467 43.714 0.2247 48.901 0.3093
L 9.247 0.0480 16.771 0.0927 18.137 0.1174
2
AA (rad/s ) N 10.241 0.0571 16.814 0.0930 18.351 0.1191
H 11.107 0.0653 16.838 0.0931 18.503 0.1201
L:18016; N: : 22520; H: 27024
133
4.3.4 Influence of suspension stiffness
The influence of the suspension stiffness on the vehicle performance under sinusoidal excitations
is performed. Also, the results are presented in terms of the time and frequency domains.
Figure 4.14 (a, b) demonstrates the time and frequency responses of the BA of the vehicle that
negotiates a shallow sinusoidal road profile with different speeds. As shown in the figure, the higher
suspension stiffness gives reasonably bigger acceleration levels. This is due to the higher energy
levels that are transmitted to the hull mass. The absolute values of the FFT responses of the BA
are also shown in Figure 4.14(b) which shows peak amplitudes at the excitation frequency of 0.55
Hz and the wheel bounce frequency of 12.5 Hz at speed of 10 km/h. For speeds of 40 km/h and 60
km/h, the peak amplitudes are due to the excitation frequencies of 2.22 Hz and 3.33 Hz and the
wheel bounce frequency of 12.5 Hz. The BA peak amplitudes are increased when the suspension
stiffness is increased. Moreover, the wheel bounce resonant frequencies are shifted up when the
suspension has higher stiffness. The PTP values of the BA responses with different values of
suspension stiffness at various speeds are listed in Table 4.15. It is seen that the PTP values of the
BA is increased when the suspension stiffness is increased. The RMS values of the system
responses are also listed in Table 4.15. Significant increases in the RMS values of the system
response occur at the bounce resonant frequency at all speeds when the suspension stiffness is
increased.
the same sinusoidal excitation. The results demonstrate that softer suspension gives a reduction in
the AA responses at all speeds. In Figure 4.15(b), the suspension has peak amplitudes at the
excitation frequencies of 0.55, 2.22 and 3.33 Hz and the wheel bounce frequency of 12.5 Hz. Also,
the figure shows the variations in the suspension stiffness affect the pitch and wheel resonant
frequencies. The stiffer suspension increase the peak amplitude in the vicinity of the pitch resonant
frequency at all speeds. Also, the softer suspension stiffness gives lower bounce resonant
frequency for the wheels. Table 4.15 shows the PTP values of the AA responses for different
suspension stiffness levels. The results show the increase in the PTP values due suspension
stiffness increase. The RMS values of the AA responses under the shallow sinusoidal excitation
are also summarized in Table 4.15.
Table ‎4.15 Effect of suspension stiffness on PTP and RMS values under shallow sinusoidal
excitation
Response stiffness (N/m) V=10 V=40 V=60
L 9.804 0.0177 11.646 0.0498 16.449 0.0996
2
BA (m/s ) N 9.826 0.0203 12.222 0.0534 16.925 0.1000
H 9.844 0.0228 12.816 0.0616 17.976 0.1034
L 15.102 0.0253 21.439 0.1271 24.055 0.1642
AA
2 N 15.467 0.0258 21.922 0.1320 24.302 0.1649
(rad/s )
H 15.827 0.0261 22.432 0.1374 24.552 0.1658
L:83200 N/m; N: : 104000 N/m; H: 124800 N/m
134
Speed 10 km/h
8 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.2
6
1
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
0
0.4
-2
0.2
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
8 4
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 3.5 High: 27024 Ns/m
6
3
4
2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2
0
1.5
-2
1
-4
0.5
-6 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
10 7
Low: 18016 Ns/m Low: 18016 Ns/m
8
6
6
4 5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
4
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.14 Influence of varying suspension stiffness on BA responses under shallow sinusoidal
road profile
135
Speed 10 km/h
14 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
12 High: 27024 Ns/m High: 27024 Ns/m
1.2
10
1
8

0.8
6
4 0.6
2
0.4
0.2
-2
-4 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
15 9
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 8 High: 27024 Ns/m
10
7
5 6
5
0
4
-5 3
2
-10
1
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
15 12
Low: 18016 Ns/m Low: 18016 Ns/m
10 10
5 8
0 6
-5 4
-10 2
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.15 Influence of varying suspension stiffness on AA responses under shallow sinusoidal
road profile
136
over medium sinusoidal profile. Figure 4.16(a, b) show the BA of the suspension responses in time
and frequency domains. The time history of the BA shown in Figure 3.21(a) demonstrates that the
stiffer suspension gives significant increase in the acceleration levels at low and medium speeds
compared with the shallow sinusoidal road. However, at high speed the BA responses are slightly
reduced when the suspension stiffness is increased. In addition, the BA spectra shown in
Figure 4.16(b) are the FFT responses at speeds of 10 km/h, 40 km/h and 60 km/h which show
peak amplitudes at the excitation frequencies of 1.11 Hz, 4.44 Hz and 6.66 Hz, respectively. They
also show peak amplitudes at the wheel bounce frequency which varies between 12.5 Hz and 13.0
Hz depending on the suspension stiffness. Table 4.16 shows that the increase in the suspension
stiffness offers high PTP values of the BA responses. The RMS values of the BA listed in
Table 4.16 show the reduction in the RMS peak amplitudes of the suspension responses at
frequencies.
shown from the figure that the AA responses are increased when the suspension stiffness is
increased. The FFT values of the AA responses shown in Figure 4.17(b) show peak amplitudes at
the excitation frequencies of 1.11 Hz, 4.44 Hz and 6.66 Hz, which correspond to the three vehicle
speeds. Also, peak amplitudes are seen at the wheel bounce frequencies which depend on the
suspension stiffness. . As the vehicle speed is increased, the PTP values of the AA are also
increased as can be seen from Table 4.16.
Table ‎4.16 Effect of suspension stiffness on PTP and RMS values under medium sinusoidal
excitation
L 13.681 0.0654 54.561 0.3575 57.530 0.3700
2
BA (m/s ) N 13.861 0.0639 55.108 0.3533 57.507 0.3633
H 14.449 0.0620 55.628 0.3496 57.483 0.3572
L 2.913 0.0158 2.421 0.0124 3.237 0.0170
2
AA (rad/s ) N 3.466 0.0187 2.495 0.0125 3.578 0.0181
H 3.983 0.0214 2.627 0.0130 3.897 0.0193
L:83200 N/m; N: : 104000 N/m; H: 124800 N/m
The change in the suspension stiffness and its effect on the vehicle performance is further
investigated under sharp sinusoidal excitation. The time and frequency domains of the BA
responses are shown in Figure 4.18(a, b). The time histories of the BA show the significant
increase in the suspension response due to high suspension stiffness. This is due to the high
energy transmitted to the vehicle hull. The FFT plots of the BA show peak amplitudes around the
excitation frequencies of 2.77 Hz, 11.11 Hz and 16.66 Hz at speeds of 10 km/h, 40 km/h and 60
km/h, respectively. Other peak amplitudes are seen at the wheel resonance frequency of between
137
12.5 Hz and 13.0 Hz depending on the suspension stiffness. The increase in the BA responses can
be noticed from the increase in the PTP values as shown in Table 4.17.
Similarly, the AA responses under the same excitation are shown in Figure 4.19(a, b).
Figure 4.19(a) shows the time history of the AA responses. The figure indicates the improvement in
the suspension responses at low speed due to the increase in the suspension stiffness. However,
at speeds 40 km/h and 60 km/h, the suspension responses are increased when the suspension
stiffness is increased. The peak amplitudes of the AA spectra shown in Figure 4.19(b) are due to
the excitation frequencies of 2.77 Hz, 11.11 Hz and 16.66 Hz caused by speeds of 10 km/h and 40
km/h, respectively. The figure also shows the peak amplitudes due to the varying wheel bounce
frequency. Table 4.17 summarizes the RMS values of the AA responses at different speeds.
Table ‎4.17 Effect of suspension stiffness on PTP and RMS values under sharp sinusoidal
excitation
L 31.817 0.2040 43.401 0.2259 46.259 0.2987
2
BA (m/s ) N 32.654 0.2051 43.568 0.2240 48.448 0.3067
H 33.597 0.2069 43.875 0.2232 50.850 0.3154
L 10.244 0.0579 16.642 0.0940 17.654 0.1170
AA
2 N 10.241 0.0571 16.814 0.0930 18.351 0.1191
(rad/s )
H 10.248 0.0564 16.962 0.0923 19.124 0.1215
L:83200 N/m; N: : 104000 N/m; H: 124800 N/m
138
Speed 10 km/h
8 4.5
Low: 18016 Ns/m Low: 18016 Ns/m
6 High: 27024 Ns/m 4 High: 27024 Ns/m
3.5
4
3
Bounce Acc (m/s2) 2
Bounce Acc (m/s2)

2.5
0
2
-2
1.5
-4
1
-6 0.5
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
30 25
Low: 18016 Ns/m Low: 18016 Ns/m
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2) 15
10
-10
5
-20
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
30 30
Low: 18016 Ns/m Low: 18016 Ns/m
20 25
10 20
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 15
-10 10
-20 5
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.16 Influence of varying suspension stiffness on BA responses under medium sinusoidal
road profile
139
Speed 10 km/h
2 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.5 High: 27024 Ns/m High: 27024 Ns/m
1.2
1
1

0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
1.5 0.8
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 0.7 High: 27024 Ns/m
1
0.6
0.5
0.5
0 0.4
0.3
-0.5
0.2
-1
0.1
-1.5 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
2 1.4
Low: 18016 Ns/m Low: 18016 Ns/m
1.5 High: 27024 Ns/m High: 27024 Ns/m
1.2
1
1
0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.17 Influence of varying suspension stiffness on AA responses under medium sinusoidal
road profile
140
Speed 10 km/h
20 14
Low: 18016 Ns/m Low: 18016 Ns/m
15 High: 27024 Ns/m High: 27024 Ns/m
12
10
10
Bounce Acc (m/s2) 5
Bounce Acc (m/s2)

8
6
-5
4
-10
2
-15
-20 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
25 15
Low: 18016 Ns/m Low: 18016 Ns/m
20
15
10
10
Bounce Acc (m/s2)
5 Bounce Acc (m/s2)
-5
5
-10
-15
-20
-25 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
30 25
Low: 18016 Ns/m Low: 18016 Ns/m
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
15
10
-10
5
-20
-30 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure ‎4.18 Influence of varying suspension stiffness on BA responses under sharp sinusoidal road
profile
141
Speed 10 km/h
6 4
Low: 18016 Ns/m Low: 18016 Ns/m
High: 27024 Ns/m 3.5 High: 27024 Ns/m
4
2

2.5
0 2
1.5
-2
-4
0.5
-6 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 7
Low: 18016 Ns/m Low: 18016 Ns/m
8
6
6
4 5
2
4
3
-2
-4 2
-6
1
-8
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14
Speed 60 km/h
10 9
Low: 18016 Ns/m Low: 18016 Ns/m
8 8
6
7
4
6
2
5
0
4
-2
3
-4
2
-6
-8 1
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure 4.19 Influence of varying suspension stiffness on AA responses under sharp sinusoidal road
profile
142
4.3.5 Influence of wheel stiffness
In this section, the effect of varying the wheel stiffness on the vehicle performance under sinusoidal
road profiles is presented. The results from the simulation are presented for the different sinusoidal
excitations in time and frequency domains.
Figure 4.20(a, b) shows the time and frequency responses of the BA of the vehicle that crosses a
shallow sinusoidal road with various speeds. The peak amplitudes of the FFT plots are due to the
excitation frequencies as described previously. It is seen that the magnitudes of these peak
amplitudes are independent of wheel stiffness at speeds of 10 km/h and 60 km/h. But at the speed
of 40 km/h, these peak amplitudes decrease as the wheel stiffness increases. This is due to the
closeness of the excitation frequency of 2.22 Hz to the bounce frequency of 1.5 Hz. The PTP and
RMS values of the BA responses at different speed are listed in Table 4.18. At - 10 km/h, both PTP
and RMS increase with wheel stiffness but at 60 km/h both decrease with wheel stiffness. On the
other hand, at 40 km/h, PTP increases while RMS decreases as wheel stiffness increases.
Similarly, Figure 4.21(a, b) shows the time and frequency domains of the AA responses under the
same excitation. The figure shows slight increases in the AA amplitudes at the excitation
frequencies of 0.55 Hz, 2.22 Hz and 3.33 Hz when the wheel stiffness is increased. These
frequencies correspond to vehicle speeds of 10 km/h, 40 km/h and 60 km/h respectively. The PTP
and RMS values of the AA responses, which are are listed in Table 4.18, increase as the wheel
stiffness increases.
Table ‎4.18 Influence of variation in wheel stiffness on PTP and RMS values under shallow
sinusoidal excitations
L 8.210 0.0195 11.405 0.0643 17.258 0.1019
2
BA (m/s ) N 9.826 0.0203 12.222 0.0534 16.925 0.1000
H 11.327 0.0209 13.219 0.0463 16.466 0.0958
L 13.348 0.0254 19.608 0.1271 21.992 0.1502
AA
2 N 15.467 0.0258 21.922 0.1320 24.302 0.1649
(rad/s )
H 17.496 0.0262 24.347 0.1343 26.949 0.1738
L:490400 N/m; N: : 613000 N/m; H: 735600 N/m
over the medium sinusoidal profile. Figure 4.22(a, b) show the BA of the suspension responses in
time and frequency domains. The time history of the BA demonstrates that the increase in the
wheel stiffness affects the BA responses and the PTP values are increased as shown in
Table 4.19. In addition, the BA spectra shown in Figure 4.22(b) indicate no significant change in the
BA peak amplitude when the vehicle crosses the medium sinusoidal road with 10 km/h. However,
at speeds of 40 km/h and 60 km/h, the RMS of the BA responses is significantly increased when
the wheel stiffness is increased as shown in Table 4.19.
143
Speed 10 km/h
8 1.4
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
1.2
6
1
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
0
0.4
-2
0.2
-4 0
0 2 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 4
Low: 490400 N/m Low: 490400 N/m
8 High: 735600 N/m 3.5 High: 735600 N/m
6 3
4 2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2 2
0 1.5
-2 1
-4 0.5
-6 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
10 7
Low: 490400 N/m Low: 490400 N/m
8
High: 735600 N/m High: 735600 N/m
6
6
4 5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
4
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.20 Influence of varying wheel stiffness on BA responses under shallow sinusoidal road
profile
144
Speed 10 km/h
16 1.4
Low: 490400 N/m Low: 490400 N/m
14
High: 735600 N/m High: 735600 N/m
1.2
12
10 1

8
0.8
0.6
4
2 0.4
0
0.2
-2
-4 0
0 2 4 0 2 4 6 8 10 12 14
Speed 40 km/h
15 9
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m 8 High: 735600 N/m
10
7
5 6

5
0
4
-5 3
2
-10
1
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
15 12
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
10 10
5 8
0 6
-5 4
-10 2
-15 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.21 Influence of varying wheel stiffness on AA responses under shallow sinusoidal road
profile
145
shown from the figure that the AA responses are reduced at speeds of 10 km/h and 40 km/h when
the wheel stiffness is increased. When the vehicle speed is further increased, the PTP values of
the AA are increased as seen from Table 4.19. The FFT values of the AA responses shown in
Figure 4.23(b) reveal that as the stiffness is increased, the peak amplitude is reduced at low and
medium speeds as seen from Table 4.19.
Table ‎4.19 Influence of variation in wheel stiffness on PTP and RMS values under medium
sinusoidal excitations
L 13.081 0.0640 46.463 0.2991 45.885 0.2875
2
BA (m/s ) N 13.861 0.0639 55.108 0.3533 57.507 0.3633
H 15.692 0.0639 60.653 0.3906 66.985 0.4291
L 3.501 0.0190 2.640 0.0129 3.472 0.0177
2
AA (rad/s ) N 3.466 0.0187 2.495 0.0125 3.578 0.0181
H 3.444 0.0185 2.392 0.0119 3.549 0.0184
L:490400 N/m; N: : 613000 N/m; H: 735600 N/m
The change in the wheel stiffness and its effect on the vehicle performance is further investigated
under sharp sinusoidal excitation. The time and frequency domains of the BA responses are shown
in Figure 4.24(a, b). The BA time histories show the increase in the BA suspension response due
to high wheel stiffness. This is due to the high rigidity of the road wheels which transmits the
shocks to the vehicle hull. The increase in the BA responses can be noticed from the increase in
the PTP values as shown in Table 4.20. The FFT amplitudes of the BA show the same increase in
the peak amplitudes around the excitation frequencies as seen from the same table.
Similarly, the AA responses under the same excitation are shown in Figure 4.25(a, b). The time
histories show the increase in the suspension responses at all speeds due to the increase in the
suspension stiffness. In addition, the AA spectra show increase in the peak amplitude at excitation
frequencies when the stiffness is increased as shown in Table 4.20.
Table ‎4.20 Influence of variation in wheel stiffness on PTP and RMS values under sharp sinusoidal
excitations
L 31.148 0.1952 78.699 0.3986 37.284 0.2362
2
BA (m/s ) N 32.654 0.2051 43.568 0.2240 48.448 0.3067
H 33.002 0.2082 50.479 0.2901 62.680 0.3866
L 8.856 0.0522 19.106 0.1024 14.179 0.0926
2
AA (rad/s ) N 10.241 0.0571 16.814 0.0930 18.351 0.1191
H 11.527 0.0606 19.614 0.1202 23.152 0.1482
L:490400 N/m; N: : 613000 N/m; H: 735600 N/m
146
Speed 10 km/h
10 4
Low: 490400 N/m Low: 490400 N/m
8 High: 735600 N/m 3.5 High: 735600 N/m
6
3
4
Bounce Acc (m/s2) 2.5
Bounce Acc (m/s2)

2
2
0
1.5
-2
1
-4
-6 0.5
-8 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
30 30
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
20
25
10
20
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0
15
-10
10
-20
5
-30
-40 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
40 30
Low: 490400 N/m Low: 490400 N/m
30 High: 735600 N/m High: 735600 N/m
25
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 15
-10
10
-20
5
-30
-40 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎4.22 Influence of varying wheel stiffness on BA responses under medium sinusoidal road
profile
147
Speed 10 km/h
2 1.4
Low: 490400 N/m Low: 490400 N/m
1.5 High: 735600 N/m High: 735600 N/m
1.2
1
1

0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
1.5 0.8
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m 0.7 High: 735600 N/m
1
0.6
0.5
0.5
0 0.4
0.3
-0.5
0.2
-1
0.1
-1.5 0
0 1 2 3 0 2 4 6 8 10 12 14
Speed 60 km/h
2 1.4
Low: 490400 N/m Low: 490400 N/m
1.5 High: 735600 N/m High: 735600 N/m
1.2
1
1
0.5
0.8
0.6
-0.5
0.4
-1
0.2
-1.5
-2 0
0 1 2 3 0 2 4 6 8 10 12 14
(a) (b)
Figure 4.23 Influence of varying wheel stiffness on AA responses under medium sinusoidal road
profile
148
Speed 10 km/h
20 14
Low: 490400 N/m Low: 490400 N/m
15 High: 735600 N/m High: 735600 N/m
12
10
10
Bounce Acc (m/s2) 5
Bounce Acc (m/s2)

8
6
-5
4
-10
2
-15
-20 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
40 25
Low: 490400 N/m Low: 490400 N/m
30 High: 735600 N/m High: 735600 N/m
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2) 15
10
-10
-20
5
-30
-40 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
Speed 60 km/h
40 30
Low: 490400 N/m Low: 490400 N/m
30 High: 735600 N/m High: 735600 N/m
25
20
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 15
-10
10
-20
5
-30
-40 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure ‎4.24 Influence of varying wheel stiffness on BA responses under sharp sinusoidal road
profile
149
Speed 10 km/h
8 4
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m 3.5 High: 735600 N/m
6
3
4

2.5
2
0
1.5
-2
1
-4
0.5
-6 0
0 1 2 3 4 0 2 4 6 8 10 12 14
Speed 40 km/h
10 9
Low: 490400 N/m Low: 490400 N/m
8 8
High: 735600 N/m High: 735600 N/m
6
7
4
6
2
5
0
4
-2
3
-4
2
-6
-8 1
-10 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
Speed 60 km/h
15 12
Low: 490400 N/m Low: 490400 N/m
High: 735600 N/m High: 735600 N/m
10 10
5 8
0 6
-5 4
-10 2
-15 0
0 0.5 1 1.5 2 0 2 4 6 8 10 12 14 16 18
(a) (b)
Figure 4.25 Influence of varying wheel stiffness on AA responses under sharp sinusoidal road
profile
150
4.4 Summary
In this chapter, the dynamic performance of the passive suspension is investigated under different
sinusoidal road profiles. The investigation includes the evaluation of suspension performance with
different configurations such as number and locations of the damper in the suspension. The
influences of different suspension design parameters are studied. This includes the effect of
damping, suspension stiffness and wheel stiffness on the suspension performance under various
road conditions. The simulation results reveal that the locations and number of dampers have a
significant effect on the suspension performance over sinusoidal terrains. The best suspension
configurations that offer optimal ride are using two or three dampers in the suspension. The best
locations of these dampers are at the first, second and last wheel stations. In addition, the passive
damper in the suspension has fixed characteristics which limit the suspension performance over
rough sinusoidal roads. Therefore, the suspension performance is reduced when the damping is
increased. Furthermore, the suspension with low stiffness offers lower acceleration responses than
the stiffer suspension. Moreover, the wheel stiffness affects the suspension dynamic behaviour.
When the wheel stiffness is increased, the suspension ride performance is reduced. These results
match the outcomes from the simulation of the passive suspension over bump terrains discussed in
chapter 3. Therefore, an improved tracked vehicle suspension with MR dampers is required. This
suspension should enhance the dynamic behaviour of the tracked vehicle over different terrains.
151
5 CHAPTER 5: EXPERIMENTAL IDENTIFICATION OF HYDRAULIC, HYDRO-GAS AND
MAGNETORHEOLOGICAL DAMPER CHARACTERISTICS
5.1 Introduction
This chapter aims to experimentally evaluate the dynamic characteristics of the conventional
hydraulic, hydro-gas and MR dampers. This evaluation provides a comparison between the
controllable dynamic characteristics of the MR dampers and uncontrollable properties of hydraulic
and hydro-gas dampers. This purpose is achieved through two stages; the first stage includes
comprehensive measurements of the three dampers dynamic performance using electro-servo-
hydraulic (ESH) testing machine.The second stage contains a restoring force surface (RFS)
technique to identify the degree of the nonlinearities inherent in the dampers. In the first stage the
force-velocity diagram or the so-called ‘hysteresis’ curve, besides the force-displacement curve
(work diagram), are used to represent the variation of the damping force with respect to the relative
displacement and velocity. The measurement of the damper characteristics is carried out under a
wide range of excitation frequencies and different dynamic amplitudes up to 12 mm. For the MR
damper characteristics, different values of the current drivers are chosen to cover all the range of
the active state of the damper. The RFS technique is then produced using a validated Chebyshev
polynomial approximation of the measured characteristics of the dampers for identification of their
dynamic performance. Moreover, a modified Bouc-Wen parametric model of the MR damper is
presented. This is followed by a validation between the predicted response of the model and the
measured characteristics of the MR damper. Furthermore, the influence of the model parameters
on the predicted response is investigated. Finally, the best set of parameters to use in the modified
Bouc-Wen model is derived to characterise the MR damper behaviour.
5.2 Experimental test setup

The electro-servo-hydraulic (ESH) testing machine used for measuring the dynamic characteristics
of the three dampers is shown in Figure 5.1. The machine consists mainly of three parts: the
loading frame, the control unit and the hydraulic power supply (not shown). The loading unit has
two vertical columns with adjustable upper crosshead and lower crosshead (fixed to the frame). In
the upper crosshead frame there is a fixture to clamp the upper end of the damper. From the upper
side of this crosshead frame a servo-actuator is mounted. The actuator has a built-in linear voltage
displacement transducer (LVDT) to measure the displacement of the damper moving end. The
lower crosshead frame also has a clamp to fix the lower end of the damper. A load cell is mounted
in this lower crossbeam for measuring the generated damping force. The control unit here is used
to adjust the frequencies and amplitudes of the excitation. The unit has two channels to measure
the displacement and force signals which can be directed to a computer through a data acquisition
card. In the case of the MR damper test, the current driver needed for activating the MR fluid is
provided by a 12 V DC power supply. The ESH machine is used to measure the dynamic
characteristics of the three different types of dampers, namely: hydraulic, hydro--gas and MR
damper. The first two dampers are used in a Citroën car while the MR damper is from Lord
152
Corporation. The machine is used to test each damper independently with the predetermined
excitation conditions.
1- Upper cross-head beam 2- Lower cross-head beam

3- Actuator drive rod 4- Damper
5- Actuator 6- Load cell
7- Control unit 8- Output signal ports
9- Data acquisition card 10- Computer
Figure ‎5.1 ESH testing machine with experimental setup
The stroke of the three dampers was measured before installation to the machine and a midpoint
which equals to half of the stroke is marked on the piston rod. This will help to adjust the piston to
its mid point to prevent the damper from damage. All dampers were excited by sinusoidal
excitations at a wide range of frequencies and amplitudes. This range represents the loading
operational conditions of the dampers. The tests are performed with constant amplitudes of
excitations ranging from 2.5 mm to 12 mm. Choosing 12 mm as a maximum amplitude value since
the maximum stroke of the actuator rod is restricted to +/-12 mm. For the MR damper active state
test, the damper is tested when the coil is energised with a range of current from 0.5 - 2 Amp.
Different combinations of frequency of 1, 2, 3, 6, 8, 11, 14 and 17 Hz and stroke of 2.5, 5, 7.5,10
and 12 mm are used during measurements. These ranges give 40 combinations. The same
153
combinations are repeated at 0.5, 1, 1.5 and 2 Amp during testing of the MR damper in its active
state.
5.3 Analysis of the measured characteristics of dampers

In this section, the measured data for the tested dampers are presented. Firstly, the time history of
the force, velocity and displacement of each damper are plotted for lower and higher amplitudes
and a fixed excitation frequency. This is followed by the presentation of selected diagrams for the
force-velocity and force-displacement under 1, 6 and 11 Hz excitations and for dynamic amplitudes
up to 12 mm. Finally, the variation of the damping force with the relative displacement and velocity
are reported under a fixed amplitude (10 mm) and different frequencies range.
5.3.1 Analysis of the measured characteristics of hydraulic damper

The measured displacement, velocity and force time histories at excitation frequency 1 Hz and two
amplitudes 2.5 and 12 mm are shown in Figure 5.2 (a, b). The figure verifies the shape of the
sinusoidal excitation of the displacement and velocity signals. The damper has two different
strokes: extension and compression. In the figure shown, the positive values of displacement,
velocity and damping force denote extension of the damper (the piston moves away from the
cylinder body. However, the compression of the damper (the piston moves toward the cylinder
body) is indicated by negative displacement, velocity and force. It is found from Figure 5.2 (a) that
the maximum damping force at comparatively small amplitudes are 102 N and -50 N in extension
and compression stroke respectively. Similarly, Figure 5.2 (b) illustrates the maximum damping
force of the damper at reasonably large amplitudes of 381 N and -282 N in extension and
compression strokes, respectively. Hence, the damper has a high damping in the extension stroke
and a low damping in the compression stroke.
0.12 0.4
Displacement Displacement
Velocity Velocity
0.1
Force 0.3 Force
0.08
Disp (m), Vel (m/s), Force (kN)
0.2
0.06
0.04 0.1
0.02 0
0
-0.1
-0.02
-0.2
-0.04
-0.06 -0.3
0 1 2 3 4 0 1 2 3 4
Time (s) Time (s)
(a) (b)
Figure ‎5.2 Displacement, velocity and force time history of hydraulic damper under sinusoidal
excitation 1Hz and (a) amplitude 2.5 mm (b) amplitude12 mm
The force-displacement (f-d) and force-velocity (f-v) characteristics of the hydraulic damper are
shown in Figure 5.3 to Figure 5.5. The figures show the characteristics of the damper at different
frequencies of 1, 6 and 11 Hz respectively and at different excitation amplitudes 2.5, 5, 7.5, and 12
mm. It is apparent from Figure 5.3 (a, b) that at relatively low frequency the damping force
154
generated by the damper ranged from -282 N to 382 N in compression and extension, respectively.
Similarly, the damper characteristics at 6 Hz and at the same range of amplitudes are depicted in
Figure 5.4 (a, b) which shows an increase in the damping force with increasing frequency. The
range of the produced damping force is from -595 N to 1760 N. Also, the characteristics of the
damper at relatively high frequency of 11 Hz are shown in Figure 5.5 (a, b). The figure shows the
increased range of damping force to be -722 N in compression and 2220 N in extension. Generally,
it can be concluded from the results that the hydraulic damper has a stiffer damping in extension
than in compression. The force-velocity characteristics of the damper evaluated at different range
of frequencies 1, 3, 6 and 8 Hz and fixed amplitude of 10 mm is shown in Figure 5.6. It can be seen
from the figure that as the excitation frequency increased, the damping force produced by the
damper is reduced in both compression and extension at a fixed velocity.
0.4 0.4
2.5 mm
5 mm
0.3 7.5 mm 0.3
10 mm 2.5 mm
12 mm 5 mm
0.2 0.2 7.5 mm
10 mm
12 mm
Force (kN)
Force (kN)
0.1 0.1
0 0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08
Displacement (m) Velocity (m/s)
(a) (b)
Figure ‎5.3 Force-displacement and force-velocity dependence of the hydraulic damper under
excitation frequency 1 Hz and dynamic amplitudes 2.5 mm to 12 mm
2 2
2.5 mm
5 mm
7.5 mm
1.5 1.5 2.5 mm
10 mm
12 mm 5 mm
7.5 mm
1 1 10 mm
12 mm
Force (kN)
Force (kN)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
155
2.5 2.5
2.5 mm
5 mm
2 7.5 mm 2
10 mm 2.5 mm
12 mm 5 mm
1.5 1.5 7.5 mm
10 mm
12 mm
Force (kN)
Force (kN)
1 1
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
(a) (b)
2 2
1 Hz
3 Hz
6 Hz
1.5 1.5 1 Hz
8 Hz
3 Hz
6 Hz
1 1 8 Hz
Force (kN)
Force (kN)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.012 -0.008 -0.004 0 0.004 0.008 0.012 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
Figure 5.6 Force-velocity characteristics for hydraulic damper at 10 mm amplitude and different
excitation frequencies
5.3.2 Analysis of the measured characteristics of hydro-gas damper

Similarly, the measured displacement, velocity and force time histories of the hydro-gas damper at
excitation frequency 1 Hz and two amplitudes, 2.5 mm and 12 mm are shown in. Figure 5.7(a, b).
The figure shows the displacement and velocity response histories of the hydro-gas damper due to
sinusoidal excitation. It is found from Figure 5.7(a) that the maximum damping force at
comparatively small amplitude are 118 N and -50 N in extension and compression stroke
respectively. Similarly, (b) illustrates the maximum damping force of the damper at reasonably
large amplitudes to be 392 N and -210 N in extension and compression stroke, respectively.
Hence, the hydro-gas damper has high damping in extension stroke and low damping in
compression stroke just similar to hydraulic damper.
The force-displacement (f-d) and force-velocity (f-v) characteristics of the hydro-gas damper are
shown in Figure 5.8 to Figure 5.10. The figures show the characteristics of the damper at the same
range of frequencies used for the hydraulic damper excitation, namely: 1, 6 and 11 Hz at different
excitation amplitudes 2.5, 5, 7.5 and 12 mm. It is apparent from Figure 5.8 (a,b) that at relatively
156
low frequency the damping force generated by the hydro-gas damper ranged from -210 N in
compression to 392 N in extension. Similarly, the damper characteristics at 6 Hz and at the same
range of amplitudes are depicted in Figure 5.9 (a, b) which shows an increase in the damping force
with increasing frequency. The range of the damping force produced is from -559 N to 1700 N.
Also, the characteristics of the damper at relatively high frequency of 11 Hz are shown in
Figure 5.10 (a, b). The figure shows the increased range of damping force to be -700 N in
compression to 2100 N in extension. Generally, it can be concluded from the results that the hydro-
gas damper has a higher damping in extension than in compression just like the hydraulic damper
but with low value of damping force.
The force-velocity characteristics of the damper evaluated at different frequencies of 1, 3, 6 and 8

Hz and fixed amplitude of 10 mm are shown in Figure 5.11. It can be observed from the figure that
the damping force produced by the damper reduced when the excitation frequency increased.
0.12 0.4
Velocity Velocity
0.1
Force 0.3 Force
0.08

0.2
0.06
0.04 0.1
0.02 0
0
-0.1
-0.02
-0.2
-0.04
-0.06 -0.3
0 1 2 3 4 0 1 2 3 4
Time (s) Time (s)
(a) (b)
Figure ‎5.7 Displacement, velocity and force time history of hydro-gas damper under sinusoidal
excitation at 1 Hz and a- 2.5 mm b- 12 mm amplitudes
0.4 0.4
2.5 mm 2.5 mm
5 mm 5 mm
0.3 7.5 mm 0.3 7.5 mm
10 mm 10 mm
12 mm 12 mm
0.2 0.2
Force (kN)
Force (kN)
0.1 0.1
0 0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08
(a) (b)
Figure ‎5.8 Force-displacement and force-velocity dependence of the hydro-gas damper under
157
2 2
2.5 mm 2.5 mm
5 mm 5 mm
7.5 mm 7.5 mm
1.5 1.5
10 mm 10 mm
12 mm 12 mm
1 1
Force (kN)
Force (kN)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
2.5 2.5
2.5 mm 2.5 mm
5 mm 5 mm
2 7.5 mm 2 7.5 mm
10 mm 10 mm
12 mm 12 mm
1.5 1.5
Force (kN)
Force (kN)
1 1
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
(a) (b)
2 2
1 Hz 1 Hz
3 Hz 3 Hz
6 Hz 6 Hz
1.5 1.5
8 Hz 8 Hz
1 1
Force (kN)
Force (kN)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-0.012 -0.008 -0.004 0 0.004 0.008 0.012 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
Figure ‎5.11 Force-velocity characteristics for hydro-gas damper at 10 mm amplitude and different
excitation frequencies
158
5.3.3 Analysis of the measured characteristics of MR damper
The MR damper characteristics are experimentally evaluated by measuring the response of the
damper at different frequencies, amplitudes and applied current. The excitation frequencies are 1,
2, 3, 6, 8, 11, 14 and 17 Hz while the dynamic amplitudes are 2.5, 5, 7.5,10 and 12 mm. There are
five constant currents applied to the MR damper which are 0, 0.5, 1, 1.5 and 2 A. Figure 5.12(a, b)
shows the dependence of displacement, velocity and force on time at excitation frequency 1 Hz
and two amplitudes 2.5 and 12 mm and constant current of zero and 2 Amp.
The figure shows the displacement and velocity response histories of the MR damper due to the
sinusoidal excitation. The positive values of displacement, velocity and damping force denote the
extension of the damper while compression is indicated by the negative values of the
displacement, velocity and damping force. It can be seen from the figure that the maximum
damping force produced has a small value of 94 N at relatively small amplitude and excitation 1 Hz
when the applied current equals zero. The maximum damping force at excitation 1 Hz and 12 mm
amplitude is 1.7 kN at 2 Amp.
0.1 2
Velocity Velocity
Force 1.5 Force
0.08
1
0.06
0.5
0.04 0
-0.5
0.02
-1
0
-1.5
-0.02 -2
0 1 2 3 4 0 1 2 3 4
Time (s) Time (s)
(a) (b)
Figure ‎5.12 Displacement, velocity and force time histories of MR damper under sinusoidal
excitation at 1 Hz: (a) 2.5 mm at 0 Amp (b) 12 mm at 2 Amp
The MR damper characteristics depicted by the force-displacement and force-velocity curves, at

various dynamic amplitudes and frequencies and at a fixed current value of 1 Amp are shown in
Figure 5.13 to Figure 5.15. It can be seen from Figure 5.13 that at relatively small frequency of 1
Hz, the damping force increases as the excitation amplitudes increased. The maximum damping
force in extension and compression are 1.43 kN and -1.36 kN, respectively. Similarly, Figure 5.14
shows the dependence of the damping force on the same ranges of amplitudes at relatively
medium frequency (6 Hz). The damping force also increased with increasing amplitudes and
reaches 1.85 kN and -1.83 in extension and compression strokes.
Also, the effect of increasing amplitude at relatively high frequency of 11 Hz is illustrated in

Figure 5.15. The figure shows the same trends of increasing damping force and the maximum
damping force reaches values of 2.44 kN and -2.32 kN in extension and compression respectively.
Figure 5.16 shows the dependence of the damping force on displacement and velocity at different
frequencies, fixed amplitude (10 mm) and constant current (1A). According to the figure, the
159
damping force increases with increasing frequency. The force-velocity loop in the figure reveals
nonlinearity between the force and velocity. The higher damping force occurs at higher frequency.
1.5 1.5
2.5 mm 2.5 mm
5 mm 5 mm
7.5 mm 7.5 mm
1 1
10 mm 10 mm
12 mm 12 mm
0.5 0.5
Force (kN)
Force (kN)
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08
(a) (b)
Figure ‎5.13 Measured characteristics of MR damper at 1 Hz excitation frequency and with
2 2
2.5 mm 2.5 mm
5 mm 5 mm
1.5 7.5 mm 1.5 7.5 mm
10 mm 10 mm
1 12 mm 1 12 mm
0.5 0.5
Force (kN)
Force (kN)
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-2 -2
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
2.5 2.5
2.5 mm 2.5 mm
2 5 mm 2 5 mm
7.5 mm 7.5 mm
1.5 10 mm 1.5 10 mm
12 mm 12 mm
1 1
0.5 0.5
Force (kN)
Force (kN)
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-2 -2
-2.5 -2.5
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
(a) (b)
amplitudes up to 12 mm and 1 Amp (a) force-displacement diagram and (b) force-velocity diagram
160
2.5 2.5
1 Hz 1 Hz
3 Hz 3 Hz
2 2
6 Hz 6 Hz
8 Hz 8 Hz
1.5 1.5
1 1
Force (kN)
Force (kN)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-2 -2
-0.012 -0.008 -0.004 0 0.004 0.008 0.012 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
(a) (b)
Figure ‎5.16 Measured characteristic of MR damper under different excitation frequencies,
amplitude 10 mm and 1 A (a) force-displacement diagram and (b) force-velocity diagram
5.4 Restoring force surface (RFS) mapping of damper characteristics using Chebyshev
Polynomial Fits
5.4.1 Restoring force surface (RFS) technique

It is well known that the standard hydraulic, hydro-gas and MR dampers characteristics are highly
nonlinear. Two approaches are used to characterize the inherent nonlinearities present in these
dampers. The first approach is the force-velocity and/or the force-displacement dependence. The
second one is the restoring force surface (RFS) which is a non-parametric technique for
identifications of damper nonlinearities [52, 89-93]. In the first method, the dampers are
experimentally tested and their dynamic properties are measured. In general, the damping force,
relative displacement and velocity are measured at different frequencies and over a wide range of
amplitudes. Obtaining the force-velocity dependence will enable the correlation between the
damping force and the corresponding relative velocity. This diagram manifests ‘hysteresis’ loops
which occur in a finite enclosed region within the diagram. The force-displacement correlation
normally called ‘work-diagram’ by the damper can also give the force and displacement
dependence.
In the second method, a more comprehensive technique known as ‘RFS’ will employ to the
measured data of the dampers to construct a three dimensional display of the displacement and
velocity dependence of the restoring force. In this section, the RFS is applied to characterize the
nonlinearities inherent in the conventional passive damper as well as hydro-gas and MR dampers.
The RFS principle is based on the equation of motion of a simple mass-spring-damper formulated
by the Newton’s second law of motion as follows:
(‎5.1)
In the above system, the internal restoring force which is the summation of the damper and
stiffness forces, acts to return the system to its equilibrium state due to the external excitation force
. By knowing the system parameters like the system mass m, damping coefficient c and spring
stiffness k and by measuring the acceleration response, Eq. (5.1) can be rewritten as:
161
(‎5.2)
The function can be considered in a general form for a linear system as a combination of the
damper force and spring force and the equation can be expressed as follows:
(‎5.3)
In the previous equation, the restoring force is assumed to be dependent on the relative
displacement and velocity, , Hence it can be constructed as a surface over the phase plane
( , ). Thus, the restoring force can be generated instantaneously as a function of known
displacement and velocity at sampling time. The equation that governs the restoring force at an
instant time is of the following form:
(‎5.4 )
By instantaneously measuring the displacement and velocity or even by the numerical

integration of the measured acceleration, a triplet of points ( ) are obtained at each instant.
The first two points represent the phase plane and the third one is the restoring force above those
points. It is shown later that the measured data of all the dampers are randomly distributed over the
phase plane; the restoring force can then be expressed by a Chebyshev approximation function.
5.4.2 Chebyshev polynomial fitting technique

As the dependence of on and is now known, a model can be fitted using the Chebyshev
approximation function of the following form:
(5.5)
where and are the Chebyshev polynomial basis and are the coefficients of the
polynomial.
It is noted that Chebyshev polynomials are orthogonal in a certain interval which is contained in the
rectangular grid given by [ , ]x[ , ] in the phase plane. However, the orthogonality
relationship, which is defined below, is only valid on the region [-1,1]x[-1,1]. This means that if
maximum displacement is outside the limits, that is , some of the data can be used for the
approximation but the accuracy will be low. In addition, when , the data will only cover a
small region.. For this reason, the measured data of the phase plane are normalised to be bounded
by the Chebyshev range [-1,1]x[-1,1] [94].The normalized values for the displacement and velocity
can be expressed as follows:
(‎5.6)
;
The Chebyshev polynomials of order n is defined as follows:
162
; (5.7)
and it satisfies the weighting orthogonality property
(5.8)
where
The coefficients of the Chebyshev function can be determined by the following double integral [95]:
(5.9)
It is noted that a 3D Chebyshev polynomial is required to represent the MR damper force as a

function of displacement, velocity and voltage as in [52]. However, only the two dimensional
orthogonal Chebyshev polynomial as in [25] is used in this research because the main aim is to
model and compare the force-displacement-velocity characteristics of the hydraulic, hydro-gas and
MR dampers rather than to model solely the force-displacement-velocity-voltage properties of the
MR dampers.
Now, the Chebyshev approximation, with the associated coefficients, is used to produce the
restoring force surface using the following steps:
Step 1: The measured force-displacement-velocity characteristics of the dampers are plotted in

three dimensional (3D) format at a certain excitation frequency and different amplitudes. These
measured data are represented by loops of different magnitudes and are not uniformly distributed
on the phase plane as shown in Figure 5.17.
Step 2: The 2D interpolation/extrapolation functions in Matlab are used to construct uniformly

distributed grid points (shown in black) on the trajectory plane of the measured data (shown in
blue) as shown in Figure 5.17.
Step 3: Then, the grid forces are computed at the grid points using the extrapolation functions.
Step 4: The grid forces are next used to calculate the Chebyshev approximation coefficients.
Step 5: The Chebyshev approximation function is validated with the measured data before it can be
used to produce the restoring force surface using equation (5.5).
More details about the RFS and Chebyshev polynomial approximation can be found in [52, 91-96].
163
Figure ‎5.17 Projected phase plane of hydraulic damper under excitation frequency of 2 Hz

In this section the comparison of the generated restoring force surface of the three dampers at
excitation frequency of 2 Hz, dynamic amplitudes of up to 12 mm and MR damper excitation
current of 1 Amp is presented. First, the measured data of the three dampers are plotted in the 3D
format as shown in Figure 5.18. In the figure, the generated grid forces are denoted by circles. A
two dimensional Chebyshev polynomial of degree 14 for the displacement and velocity is required
to produce the RFS along with 225 corresponding coefficients which are shown in Tables 5.1 to
5.3. The 2D interpolation function is then validated against the measured characteristics of the
dampers with amplitude of 5 mm. Figure 5.19 shows the good match between the measured force-
displacement and force-velocity and the predicted characteristics using the derived Chebyshev
polynomials.
Figure 5.20(a) shows the RFS (which are produced by the predicted Chebyshev approximation) of
the damping force of the hydraulic damper with respect to the relative displacement and velocity.
The figure shows that the dynamic characteristics of the hydraulic damper vary smoothly with the
increase of the relative velocity but with a nonlinear behaviour. Similar dynamic properties are
noticed for the hydro-gas damper however with more nonlinearity as shown in Figure 5.20(b). The
figure also reveals that the damping forces of both the hydraulic and hydro-gas dampers vary fairly
smoothly with the relative velocity. From a relatively high negative velocity to high positive velocity
the damping force varies from a small negative value to a high positive value. These variations
make both dampers to be soft in compression strokes and stiff in the extension strokes. On the
other hand, Figure 5.20(c and d) show the MR damper dynamics with two command currents of 0
and 1 Amp. It can be seen that the damping force of the MR damper is highly dependent on the
relative velocity and the applied current. The MR damping force varies smoothly at low to high
negative velocities but at low to high positive velocities its variation is sharp. The MR damper
shows a much larger and nonlinear damping force variation compared to the hydraulic and hydro-
164
gas dampers. The force variation implies that the damper is hard in the extension stroke and soft in
the compression stroke. It should be noted that the very sharp discontinuity seen in the RFS map
of the MR damper is due to the yield and plug flow characteristic of MR fluids [97-98].
The damping force of the hydraulic damper at 2 Hz ranges from -428 N to 1127 N and the hydro-
gas damper force ranges from -398 N to 1143 N. However, the MR damper damping force changes
from -81.5 N to 193.1 N at zero Amp and from -1436 N to 1518N at 1 Amp. These values imply that
the dampers are all soft in compression and stiff in extension. This feature is as required for all
dampers. In the compression stroke, the restoring force should be relatively small to softly absorb
the shocks. But in the case of the extension stroke, which is the dissipation energy process, the
restoring force needs to be very high. The MR damper restoring force follows this property and
when it is activated by an applied current of 1 A, it has the highest damping force magnitude and
range for the same range of relative velocity.

Similarly, the same procedures discussed earlier are used to produce the RFS of the dampers at 6
Hz. Figure 5.21 shows the measured data and the constructed grid forces at excitation frequency
of 6 Hz and amplitude up to 12 mm. The Chebyshev polynomials of degree 14 is generated with
the corresponding coefficients (shown in Tables 5.4 to 5.6) to produce the restoring force surface.
In addition, the 2D polynomials of the force-displacement and the force-velocity characteristics are
also validated with the measured data of the three dampers at amplitude 10 mm. Figure 5.22
shows good match between the work diagram and the characteristics diagram of the measured
data and the Chebyshev polynomial approximations. Figure 5.23 shows the produced restoring
force surfaces of the three dampers using the validated polynomial approximations. It is evident
from the figure that all the dampers have significant nonlinear characteristics. The hydraulic and the
hydro-gas dampers have smoothly varying nonlinear characteristics. The reason for the high
nonlinearity of the hydraulic and hydro-gas dampers might be because both dampers have gas
chambers filled with nitrogen gas. In the case of the MR damper characteristics shown in
Figure 5.23(c and d), it is revealed that this damper has highly nonlinear characteristics compared
to the other dampers over the same dynamic range of the relative velocity. Also, at 6 Hz frequency
the hydraulic damping force varies from -595 N to 1763 N and the hydro-gas damper has a
relatively lower damping force range from -559 N to 1697 N compared with the hydraulic damper.
However, the damping force of the MR damper ranges from -355.5 N to 517 N at zero Amp and
from -1825 N to 1851 N at 1.0 A.
165
1500
1000
Force (N)
500
-500
0.15
0.1
0.05 0.015
0 0.01
0.005
-0.05 0
-0.1 -0.005
-0.15 -0.01
-0.2 -0.015
Velocity (m/s) Displacement (m) (s)
Hydraulic damper
1200
1000
800
600
Force (N)
400
200
-200
-400
0.15
0.1
0.05 0.015
0 0.01
0.005
-0.05 0
-0.1 -0.005
-0.15 -0.01
-0.2 -0.015
Hydro-gas damper
2000
1500
1000
Force (N)
500
-500
-1000
-1500
0.15
0.1
0.05 0.015
0 0.01
0.005
-0.05 0
-0.1 -0.005
-0.15 -0.01
-0.2 -0.015
MR damper
Figure ‎5.18 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 2
Hz excitation and amplitudes up to 12 mm; measured data (solid line) and grid forces (circles)
166
400 400
Chebyshev approximation Chebyshev approximation
Experimental tests Experimental tests
300 300
200 200
100 100
Force (N)
Force (N)
0 0
-100 -100
-200 -200
-300 -300
-6 -4 -2 0 2 4 6 -0.06 -0.04 -0.02 0 0.02 0.04 0.06
Displacement (m) x 10
-3 Velocity (m/s)
Hydraulic damper
400 400
300 300
200 200
100 100
Force (N)
Force (N)
0 0
-100 -100
-200 -200
-300 -300
-6 -4 -2 0 2 4 6 -0.06 -0.04 -0.02 0 0.02 0.04 0.06
-3 Velocity (m/s)
Hydro-gas damper
1500 1500
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-6 -4 -2 0 2 4 6 -0.06 -0.04 -0.02 0 0.02 0.04 0.06
-3 Velocity (m/s)
MR damper
(a) (b)
Figure ‎5.19 Validation of experimental test data and Chebyshev approximation under excitation
frequency 2 Hz and amplitude 5 mm (a) Force-Displacement and (b) Force-Velocity
167
1100 1100
950 950
Damping Force (N)
Damping Force (N)

750 750
550 550
350 350
150 150
-50 -50
-250 -250
-450 -450
0.1 0.12
0.08
0.05 0.012 0.012
0.008 0.03 0.008
0 0.004 0.004
0 -0.02 0
-0.05 -0.004 -0.07 -0.004
-0.1 -0.008 -0.008
Velocity (m/s) -0.012 Velocity (m/s) -0.12 -0.012
Displacement (m) Displacement (m)
(a) Hydraulic Damper (b) Hydro-gas Damper
200 1800
1600
150 1200
Damping Force (N)
Damping Force (N)
800
100
400
50 0
0 -400
-800
-50
-1200
-100 -1600
0.13
0.12 0.12
0.07 0.012 0.08 0.012
0.02 0.008 0.03 0.008
0.004 -0.02 0.004
-0.03 0 0
-0.08 -0.004 -0.07 -0.004
-0.008 -0.12 -0.008
Velocity (m/s) -0.13 -0.012 Velocity (m/s) -0.012
(c) MR Damper (zero Amp) (d) MR Damper (1 Amp)

Figure ‎5.20 Comparison of restoring force surface characteristics of dampers under excitation
frequency 2 Hz, amplitude up to 12 mm
168
Table ‎5.1 Calculated Chebyshev coefficients i for hydraulic damper at excitation frequency of 2 Hz and amplitudes up to 12 mm
i
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 146.00 689.10 141.32 10.35 7.27 -8.06 -9.16 1.48 6.47 -0.07 -0.54 3.67 2.15 -0.68 -0.16
1 25.20 4.06 -16.62 9.71 -8.71 -13.92 2.83 1.09 -2.85 -4.02 -4.27 0.28 0.77 0.79 -0.33
2 -4.53 -11.44 -3.00 5.50 0.16 3.78 6.02 0.13 2.44 3.57 -0.05 -2.49 -4.07 -1.96 0.60
3 -1.77 -0.28 2.61 2.31 2.34 -0.08 -3.05 -2.56 -1.71 -0.23 3.16 1.91 -1.98 -2.18 -0.97
4 1.88 4.39 -0.02 -3.83 -1.50 0.52 -0.85 1.45 -0.41 -1.89 -2.42 -0.32 2.21 4.43 3.05
5 -0.55 -1.50 -1.09 -0.78 -0.29 0.93 1.85 0.96 1.53 0.73 -1.36 -0.36 1.00 -0.11 -1.18
6 -0.07 0.44 1.15 2.08 0.42 -2.50 -1.46 -1.21 -0.59 1.49 2.35 0.44 -1.71 -3.99 -1.70
7 -0.05 1.30 -0.98 -1.95 -0.24 0.39 0.75 2.15 0.09 -1.29 -1.73 -1.03 0.82 2.27 3.08
8 -0.94 -1.24 -0.25 0.35 1.75 1.57 -0.33 -0.47 -0.09 -0.24 0.67 0.74 -0.37 0.23 -0.69
9 0.15 -0.19 1.29 1.51 0.03 -0.20 -1.71 -1.50 0.00 0.48 0.69 0.82 -0.46 -1.43 -1.94
10 1.03 0.30 -0.27 -1.57 -2.58 -0.75 1.82 2.00 1.01 -0.52 -1.39 -1.68 0.56 1.58 1.15
11 -0.51 -0.51 0.29 1.07 2.03 -0.26 -0.58 -1.93 -1.23 0.60 2.43 0.43 -1.00 -0.62 -0.84
12 -0.24 0.35 0.01 -0.19 0.87 0.97 -0.70 -1.03 -0.57 -0.28 -0.04 1.20 1.23 0.71 0.20
13 0.66 0.31 -0.99 -1.99 -2.50 0.30 2.00 2.46 1.30 -0.67 -3.02 -0.98 1.76 1.56 1.62
14 -0.35 -0.07 1.05 2.19 -0.01 -1.62 -0.35 -1.20 -0.80 1.57 1.32 0.16 -2.31 -4.40 -1.97
169
Table ‎5.2 Calculated Chebyshev coefficients i for hydro-gas damper at excitation frequency of 2 Hz and amplitudes up to 12 mm
i
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 180.23 680.65 158.74 24.85 -6.70 -12.54 -8.80 -4.63 3.59 1.59 -2.10 1.91 6.12 2.70 -1.33
1 11.61 -13.21 -21.80 -5.96 -14.13 -7.74 -1.75 3.25 5.27 2.77 0.35 2.47 3.11 -0.16 -1.79
2 -3.04 -5.63 -1.53 0.99 0.93 0.81 1.50 0.11 0.65 2.17 1.77 -1.20 -2.49 -3.09 -0.60
3 -0.52 3.22 3.05 3.21 3.74 1.54 1.57 0.10 -3.03 -0.41 1.18 -0.82 -2.01 -2.17 0.65
4 0.86 1.91 0.21 -1.44 0.19 0.57 -1.23 -1.03 -0.39 -1.97 -1.72 0.05 1.74 3.73 0.43
5 -0.07 -0.96 -1.07 -1.42 -1.48 -1.71 -1.65 -0.85 0.69 0.36 -0.06 1.35 0.52 1.76 -0.75
6 -0.07 -0.58 0.72 1.61 -0.35 -0.12 1.36 0.70 1.04 2.72 1.37 -0.74 -1.23 -3.45 -1.15
7 -0.20 0.69 1.22 0.90 1.51 0.55 -0.05 0.31 -0.57 -0.70 0.44 0.10 0.28 -1.00 1.42
8 -0.97 -0.84 -1.17 -0.65 0.67 0.67 -0.55 -0.53 -0.57 -1.55 -1.51 0.21 0.18 1.41 0.92
9 0.55 -0.05 -0.31 -0.71 -1.14 0.17 -0.76 -0.31 0.84 -0.25 -0.47 0.13 0.23 0.78 -0.44
10 0.74 0.90 1.09 0.19 -0.70 -1.02 0.36 1.16 0.97 1.48 1.57 -0.39 -0.52 -0.99 -0.62
11 -0.53 0.27 0.10 0.00 0.41 -0.40 0.78 -0.34 -0.53 0.46 0.59 0.05 -0.26 0.22 0.02
12 -0.22 -0.36 -0.98 -0.48 0.81 1.27 -0.37 -0.78 -1.50 -2.79 -1.94 0.06 0.34 1.41 -0.11
13 0.52 -0.02 -0.08 0.15 -0.76 0.00 -0.14 0.35 0.02 0.00 -0.67 -0.02 0.86 -0.23 0.20
14 0.04 -0.34 0.46 0.12 -0.91 -0.64 1.80 1.24 1.61 3.11 1.58 0.05 -0.24 -2.13 0.15
170
Table ‎5.3 Calculated Chebyshev coefficients i for MR damper at excitation frequency of 2 Hz, amplitudes up to 12 mm and 1 Amp
i
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 21.28 1691.50 36.31 -319.98 -36.61 76.49 20.04 -6.06 -3.28 3.94 -6.47 -18.08 3.56 18.35 -1.16
1 258.16 5.50 -445.60 -32.83 289.60 58.65 -131.12 -57.51 23.34 34.96 6.46 -10.93 10.78 2.03 -38.35
2 -15.82 -66.62 27.10 130.16 -6.02 -132.03 -20.31 87.07 42.39 -26.28 -42.31 -13.45 29.39 18.28 -17.93
3 -24.39 10.54 60.92 -15.56 -72.69 14.44 85.10 -2.70 -67.66 -11.87 40.39 21.88 -6.10 -13.40 -12.15
4 -1.34 19.87 1.73 -37.25 -2.70 48.63 5.92 -54.94 -11.70 46.05 19.16 -34.07 -29.17 19.76 35.02
5 1.27 -6.40 -11.09 14.29 23.73 -15.99 -39.29 11.26 42.87 -2.50 -34.40 -4.20 20.46 3.79 -9.03
6 9.80 -3.44 -16.83 0.25 6.89 -13.65 -3.06 29.69 1.96 -39.63 -10.89 39.66 26.88 -27.45 -36.57
7 -3.98 0.31 5.22 -3.66 -0.73 4.82 3.00 -7.56 -10.54 13.60 19.83 -14.74 -30.52 6.76 36.57
8 -9.66 5.61 17.97 -5.97 -12.94 12.89 13.00 -12.66 -11.41 12.12 12.51 -9.64 -18.71 7.94 20.09
9 -1.41 1.27 1.62 -1.29 -2.87 -1.22 -0.35 6.04 4.08 -14.94 -14.16 17.16 25.67 -9.22 -35.01
10 7.72 -4.51 -15.98 7.91 18.75 -6.78 -21.27 2.62 21.44 5.55 -13.74 -9.86 10.37 7.29 -7.76
11 6.80 1.12 -9.15 -3.35 1.69 5.40 7.43 -7.12 -11.27 8.94 14.90 -8.22 -15.37 4.17 14.19
12 -8.42 -0.68 18.62 2.77 -21.08 -6.26 20.78 6.97 -20.49 -9.31 12.63 6.91 -7.66 -5.83 5.24
13 -7.01 -2.33 11.87 6.10 -3.89 -6.57 -4.44 5.35 10.58 -2.14 -13.02 0.63 8.42 -0.39 -1.05
14 12.00 4.55 -25.29 -15.38 21.08 19.54 -15.68 -19.84 13.28 10.61 -9.51 0.23 6.98 -1.12 -6.65
171
2000
1500
1000
Force (N)
500
-500
-1000
0.5
0.4
0.3 0.015
0.2
0.1 0.01
0 0.005
-0.1 0
-0.2 -0.005
-0.3
-0.4 -0.01
-0.5 -0.015
Hydraulic damper
2000
1500
1000
Force (N)
500
-500
-1000
0.5
0.4
0.3 0.015
0.2
0.1 0.01
0 0.005
-0.1 0
-0.2 -0.005
-0.3
-0.4 -0.01
-0.5 -0.015
Hydro-gas damper
3000
2000
1000
Force (N)
-1000
-2000
-3000
0.5
0.4
0.3 0.015
0.2
0.1 0.01
0 0.005
-0.1 0
-0.2 -0.005
-0.3
-0.4 -0.01
-0.5 -0.015
MR damper
Figure ‎5.21 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 6
Hz excitation and amplitudes up to 12 mm; measured data (solid line) and grid forces (circles)
172
2000 2000
1500 1500
1000 1000
Force (N)
Force (N)
500 500
0 0
-500 -500
-1000 -1000
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
Hydraulic damper
2000 2000
1500 1500
1000 1000
Force (N)
Force (N)
500 500
0 0
-500 -500
-1000 -1000
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
Hydro-gas damper
2500 2500
2000 Experimental tests 2000 Experimental tests
1500 1500
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-2000 -2000
-2500 -2500
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
MR damper
(a) (b)
Figure ‎5.22 Validation of experimental tests and Chebyshev approximation under excitation
frequency 6 Hz and amplitude 10 mm (a) Force-Displacement and (b) Force-Velocity
173
2000 2000
1800 1800
Damping Force (N)
Damping Force (N)

1400 1400
1000 1000
600 600
200 200
-200 -200
-600 -600
0.44 0.44
0.36
0.22 0.012 0.16 0.012
0.008 0.008
0 0.004 -0.04 0.004
0 0
-0.22 -0.004 -0.24 -0.004
-0.008 -0.008
Velocity (m/s) -0.44 -0.012 Velocity (m/s) -0.44 -0.012
(a) Hydraulic Damper (b) Hydro-gas Damper
600 2000
1600
400 1200
Damping Force (N)
Damping Force (N)
800
200 400
0
0 -400
-800
-200 -1200
-1600
-400 -2000
0.44 0.44
0.36 0.36
0.16 0.012 0.16 0.012
0.008 0.008
-0.04 0.004 -0.04 0.004
0 0
-0.24 -0.004 -0.24 -0.004
-0.008 -0.008
Velocity (m/s) -0.44 -0.012 Velocity (m/s) -0.44 -0.012
(c) MR Damper (zero Amp) (d) MR Damper (1 Amp)

Figure ‎5.23 Comparison of restoring force surface characteristics of damper under excitation
frequency 6 Hz, amplitude up to 12 mm
174
Table ‎5.4 Calculated Chebyshev coefficients i for hydraulic damper at excitation frequency of 6 Hz and amplitudes up to 12 mm
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 433.32 1333.93 209.56 -216.81 -99.26 70.16 68.74 -11.49 -41.42 -8.49 18.02 4.05 -10.00 -4.56 3.30
1 25.38 -7.81 -51.96 8.58 46.04 7.71 -17.51 -9.57 -1.48 9.53 7.54 -11.19 -12.58 3.85 11.19
2 -11.63 -15.51 14.91 31.24 -0.81 -23.79 -2.65 11.58 -0.42 -6.08 -0.15 0.93 0.75 2.71 1.32
3 -2.01 2.40 4.56 -1.89 -5.58 -0.06 4.34 -2.23 -6.36 0.50 3.93 1.15 1.94 0.26 -3.18
4 4.00 5.11 -3.39 -9.29 -1.59 5.48 1.06 -2.85 1.64 2.82 -0.98 -0.79 2.02 -0.48 -3.15
5 1.58 -0.07 -2.58 -2.02 0.42 1.32 -0.82 1.13 3.54 0.70 -2.91 -1.36 -1.10 0.18 1.53
6 -0.77 0.32 0.91 -0.10 -0.79 -0.63 1.89 2.62 -1.01 -2.11 0.62 0.12 -2.33 -0.36 2.61
7 -1.30 -0.54 1.55 1.62 1.06 -0.63 -2.03 -1.44 0.36 0.06 1.01 -0.26 -1.00 0.36 1.71
8 -0.23 -0.94 -0.59 0.86 2.01 1.51 -1.48 -1.60 -0.82 -1.46 -0.03 3.72 2.57 -1.43 -2.53
9 -0.12 -0.11 0.27 -0.58 -0.49 0.57 1.41 0.41 -0.87 -0.40 0.00 1.16 1.01 -0.81 -1.52
10 -0.08 0.27 0.73 -0.09 -1.77 -1.36 0.48 0.30 1.12 2.05 -0.41 -3.22 -1.48 1.05 0.93
11 1.02 0.83 -0.88 -0.02 -0.34 -1.14 0.29 0.93 -0.02 0.22 -0.08 -1.11 -0.35 0.96 0.99
12 1.33 0.32 -1.95 -0.20 1.67 0.19 -1.59 -1.02 0.67 1.44 0.49 -1.33 -0.70 1.22 1.76
13 -1.49 -0.93 1.97 1.02 -0.65 0.04 0.06 0.30 0.27 -1.22 -0.58 1.64 1.33 -1.19 -2.42
14 -1.49 -0.49 1.19 -1.19 -0.41 2.43 2.33 -0.34 -2.39 -2.06 0.10 2.41 1.12 -1.45 -2.05
175
Table ‎5.5 Calculated Chebyshev coefficients i for hydro-gas damper at excitation frequency of 6 Hz and amplitudes up to 12 mm
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 420.33 1253.01 194.53 -191.25 -84.63 67.58 64.55 -14.02 -40.39 -8.29 18.54 7.22 -8.71 -4.01 5.30
1 14.02 -20.75 -41.24 22.70 50.16 6.53 -21.34 -18.15 1.50 20.22 6.95 -21.84 -20.04 12.80 26.26
2 -13.13 -17.93 15.02 29.11 3.20 -16.52 0.03 4.99 -3.31 -6.38 -0.42 3.69 3.69 1.47 -2.43
3 -0.46 3.14 2.98 -3.24 -4.95 -0.52 4.61 -3.73 -5.09 0.01 2.48 4.05 3.41 -2.94 -5.52
4 5.41 5.76 -3.13 -10.50 -3.69 3.20 0.98 -0.38 2.79 0.72 0.31 -0.68 0.13 -0.58 -0.83
5 1.45 0.50 -0.12 -3.09 -0.98 -0.62 1.31 4.07 3.14 -1.38 -3.25 -2.61 0.06 0.00 0.16
6 -1.46 -0.61 1.64 0.58 -0.88 -1.27 0.00 1.07 1.02 0.22 -1.03 -2.78 -1.37 1.48 1.61
7 -2.84 -1.13 2.60 2.15 1.71 -0.01 -3.11 -2.70 0.41 1.74 0.96 -0.79 -0.01 1.79 1.72
8 -0.26 -1.08 -0.66 0.90 2.63 2.50 -0.60 -3.07 -2.59 -1.24 0.92 4.41 2.53 -1.96 -2.70
9 1.56 0.01 -2.70 -1.15 0.11 1.88 2.08 0.19 -1.98 -1.36 1.71 2.61 -0.99 -2.22 -0.26
10 0.93 1.40 -0.88 -0.91 -1.22 -1.22 1.36 2.52 0.89 0.02 0.01 -1.14 -1.78 0.01 1.08
11 0.60 0.85 0.28 1.18 -1.73 -2.81 -0.62 2.15 1.93 0.65 -2.28 -1.71 -0.07 1.04 0.42
12 -0.26 0.27 0.92 0.68 -1.05 -0.67 -1.91 -1.09 1.20 2.97 -1.27 -3.00 0.74 2.30 0.68
13 -1.27 -0.74 0.78 0.54 0.95 1.47 -0.03 -0.48 -1.26 -1.53 1.15 1.57 0.82 -0.25 -1.76
14 -0.64 -0.77 -1.73 -0.52 3.36 3.31 0.76 -1.10 -2.84 -1.69 2.30 3.08 -1.27 -2.16 0.95
176
Table ‎5.6 Calculated Chebyshev coefficients i for MR damper at excitation frequency of 6 Hz, amplitudes up to 12 mm and 2 Amp
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 -21.18 2582.73 12.07 -498.76 -24.44 133.46 13.00 26.73 5.01 -65.38 -18.72 38.10 32.63 -26.43 -42.37
1 336.44 -9.17 -592.04 6.51 445.93 24.40 -296.40 -38.23 141.28 45.08 -40.46 -35.50 -14.58 13.64 21.04
2 -16.85 -69.21 31.75 131.39 -16.41 -81.05 3.96 23.37 -16.00 -16.81 28.34 12.83 -42.68 6.35 45.13
3 -108.41 9.13 190.10 0.04 -152.37 -20.19 118.34 26.65 -71.78 -41.45 19.74 38.16 14.53 -19.46 -25.61
4 10.28 22.74 -20.69 -46.36 25.64 42.25 -33.30 -35.35 33.74 33.42 -26.89 -27.51 18.66 10.09 -11.49
5 44.66 3.98 -80.19 -21.41 71.47 26.87 -57.51 -22.59 40.50 30.90 -20.63 -26.86 0.69 12.93 13.18
6 -12.97 -2.08 24.66 13.85 -31.14 -31.73 38.78 40.92 -30.91 -35.83 25.14 27.51 -14.83 -14.72 6.06
7 -23.25 -8.72 43.93 26.79 -41.48 -29.31 29.49 24.89 -19.32 -20.40 15.86 10.27 -10.09 -3.00 1.11
8 22.55 2.49 -37.82 -18.71 31.01 35.12 -26.79 -33.39 15.23 19.71 -15.44 -10.40 12.33 6.63 -12.45
9 11.70 3.63 -24.37 -15.64 26.17 22.75 -23.09 -26.35 15.94 19.26 -11.61 -6.23 10.02 -1.15 -6.72
10 -25.91 -4.08 43.91 18.70 -35.14 -29.57 26.22 25.10 -14.47 -12.88 13.62 1.93 -8.19 0.30 11.02
11 -2.21 -0.88 6.22 7.82 -12.34 -13.48 23.10 18.17 -22.40 -17.61 10.42 14.71 -1.17 -6.19 -0.45
12 25.16 1.30 -44.44 -7.27 41.30 13.77 -34.09 -16.59 22.75 18.79 -17.99 -12.21 6.37 5.80 -1.17
13 2.54 1.06 -4.13 -4.98 7.24 6.68 -19.34 -7.18 20.68 9.75 -6.18 -16.48 -7.09 11.88 8.10
14 -27.70 3.06 46.67 -4.73 -41.63 1.65 35.04 4.60 -22.68 -19.63 14.17 23.77 -1.28 -18.03 -10.37
177
5.5 Modified Bouc-Wen Model of MR Damper
This section presents the generalized MR damper model which simulates the measured MR
damper dynamic characteristics. The model differential equations are presented and their
parameters are defined. The effect of the model parameters on the accuracy of the developed
model is discussed.
5.5.1 Dynamic model formulation

The mechanical model of the MR damper depicted in Figure 5.24 has been shown to accurately
simulate the performance of the MR damper over a wide range of excitations [34, 58, 98]. This
model has been employed in vibration control of vehicle suspension system equipped with two MR
dampers [34]. The model includes various springs and dashpots to simulate the damper dynamic
characteristics. The accumulator stiffness is represented by a spring with constant, , the damping
at high velocities is denoted by, . A dashpot with damping , is introduced in the model to
simulate the hysteresis loop at low frequencies. Another spring, with constant, is introduced into
the model to control the stiffness at large velocities. It is supposed that the initial displacement due
th
to the accumulator stiffness is, . The internal displacement for the i damper is governed by
the following equation [34],
θ θ- , i=1,2, ... (‎5.10)
th
where, is the evolutionary variable that represents the hysteresis loop of the i damper (
and it can be expressed as follows [34]
(‎5.11)
The vertical bounce and pitch of the vehicle body are b and θ respectively and the wheel bounce
is . The shape of the hysteresis loop can be adjusted by the constants, , , , and . From the
representation of the MR damper shown in Figure 5.24, the MR damper force can be written as:
θ (‎5.12)
It is noted that the control voltage u is responsible for altering the MR damper force. This voltage is
related to the supply voltage v by the following equations:
(‎5.13)
(‎5.14)
(‎5.15)
178
(‎5.16)
In order to precisely simulate the behaviour of the MR damper using the voltage-dependant
modified Bouc-Wen model given by the previous equations, a set of 14 constants which relate the
damper characteristics to the applied voltage, should be identified. This is achieved by an
optimization process in which a set of parameters is determined to fit the response of the Bouc-
Wen model to the measured response of the MR damper [98-99]. Table 5.7 shows the optimized
parameters for the MR model as employed in [58].
Figure ‎5.24 Schematic view of the Modified Bouc-Wen model of MR damper
Figure 5.25 Block diagram of the Modified Bouc-Wen Simulink model for MR damper
179
Table ‎5.7 Parameters of MR damper [58]
Parameters value parameters value
784 Ns/m 12441 N/m
1803 Ns/Vm 38430 N/Vm
2
3610 N/m 136320 /m
2
14649 Ns/m 2059020 /m
34622 Ns/Vm A 58
840 N/m n 2
0.0245 m 190 /s
5.5.2 Effect of model parameters on the predicted response

This section presents the comparison between the response of the modified Bouc-Wen model and
the measured characteristics of the MR damper. The predicted force-displacement and force-
velocity characteristics of the modified Bouc-Wen model are compared with the measured data of
the MR damper under sinusoidal excitation of frequency 3Hz and 6 Hz, dynamic amplitude of 10
mm and current of 1 Amp. The simulated results of the MR damper are produced using a model
which is built in the Matlab/Simulink environment. Figure 5.25 shows the Simulink model which is
developed as a direct representation of the Modified Bouc-Wen differential equations presented in
section 5.5.1. In the simulation, the optimized parameters of the MR damper listed in Table 5.7 and
new sets of parameters which have been derived in these investigations and are listed in Table 5.8
are used. It should be noted that the same excitation frequencies, amplitude and applied current
are used for the sets of parameters listed in Tables 5.7 and 5.8.
Figure 5.26 shows the comparisons between the predicted and measured force-displacement and
force-velocity characteristics of the MR damper under excitation frequency of 3 Hz, 10 mm
amplitude and 1 Amp. The parameters used are those listed in Table 5.7 and used for a quarter-car
model in [58]. It can be seen from the figure that these parameters do not produce a close match
between the predicted and measured force-displacement and force-velocity characteristics of the
MR damper. Moreover, at certain relative velocities the model produces lower damping forces than
the measured MR damper force. Similarly, Figure 5.27 shows the comparisons between the
predicted and measured MR characteristics under excitation frequency of 6 Hz and the same
amplitude and applied current. Again, this figure show that the parameters in Table 5.7 do not
produce a close match between the predicted and measured force-displacement and force-velocity
characteristics of the MR damper.
180
2000 2000
Modified Bouc-Wen Modified Bouc-Wen
Measured Measured
1500 1500
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
Figure ‎5.26 Comparison between force-displacement and force-velocity characteristics predicted

by modified Bouc-Wen and measured characteristics under excitation of 3 Hz, 10 mm amplitude
and current of 1 Amp
2500 2500
Modified Bouc-Wen Modified Bouc-Wen
2000 Measured 2000 Measured
1500 1500
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-2000 -2000
-2500 -2500
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
Figure ‎5.27 Comparison between force-displacement and force-velocity characteristics predicted

by modified Bouc-Wen and measured characteristics under excitation of 6 Hz, 10 mm amplitude
and current of 1 Amp
It is obvious that the accuracy of the MR damper model is determined by the difference between
the measured characteristics and those predicted by the model. When the response of the model
agrees closely with the measured response, the model can be used in simulation to represent the
MR damper characteristics. Therefore it is essential to determine accurate estimates of the MR
damper parameters. The determination of accurate estimates of the parameter values can be done
using systematic search methods or optimization methods with the objective to make the predicted
behaviour similar to or as close as possible to the measured response.
In order to show the effect of the identified parameters on the model response, and there, to obtain
better estimates of the damper parameters, a systematic method is employed to obtain better
estimates of the modified Bouc-Wen model parameters in Table 5.7 (reference parameters). In this
method , some of the constant values of the reference parameters are varied systematically in
181
order to predict characteristics that are similar to the measured behaviour. These parameters
control the dynamic force range and/or the shape of the damper hysteresis. By this process, four
sets of parameters were determined and they are used to fit the modified Bouc-Wen model to the
experimental data. Table 5.8 shows the reference parameters (taken from Table 5.7) and the 4
sets of derived parameters used for the fitting process.
Figure 5.28 shows the force-displacement and force-velocity characteristics of the predicted
characteristics using the four sets of parameters and the measured response of the MR damper
under sinusoidal excitation of 3 Hz and 10 mm amplitude at 1 Amp. The figure shows that the fitted
response of the model using Set1 parameters is much better than the fit using the reference
parameter values (RPV) but it did not produce a very close prediction of the measured
characteristics of the damper. The force-displacement and force-velocity behaviour of the predicted
response show lower damping force than the measured characteristics. For the response produced
by Set2 parameters, the model has a similar positive damping force as the measured response but
the model overpredicts the negative damping force. It can be seen that Set3 parameters
overpredict both the positive and the negative peak damping forces, especially near the peak
velocities. Finally, the figure shows that the predicted model response using Set4 parameters
shows good agreement with the measured damper characteristics.
Thus, it can be concluded that the MR dampers used in this study can be represented by the
modified Bouc-Wen model with Set4 parameters. The model parameters have a potential effect on
the model accuracy. There is no simple parametric model with high accuracy for the MR dampers,
Table 5.8 Sets of parameters for fitting Modified Bouc-Wen model to experimental data
MR Constants Reference values Set 1 Set 2 Set 3 Set 4
Ns/m 784 100 300 600 300
Ns/Vm 1803 1000 1000 1500 1000
N/m 3610 3610 3610 3610 3610
4 4 4
Ns/m 14649 14649 2x10 4x10 2x10
4 4 4
Ns/Vm 34622 34622 2x10 4x10 4x10
N/m 840 840 840 840 840
m 0.0245 0.0245 0.0245 0.05 0.05
4 4 4
N/m 12441 12441 3x10 10 3x10
4 4 4
N/Vm 38430 38430 4x10 4x10 4x10
-2 4 4 4 4
m 136320 10 10 10 10
-2 5 5 5 5
m 2059020 5x10 5x10 5x10 5.5x10
A 58 50 50 50 50
n 2 2 2 2 2
-1
s 190 190 190 190 190
182
2000 2000
Predicted by set1
Measured
1500 1500
1000 1000
Force (N)
Force (N)
500 500
0 0
-500 -500
-1000 -1000
-1500 -1500
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
2000 2000
Predicted by set2
1500 1500 Measured
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-2000 -2000
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
2000 2000
Predicted by set3
1500 1500 Measured
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-2000 -2000
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
2000 2000
Predicted by set4
1500 1500 Measured
1000 1000
500 500
Force (N)
Force (N)
0 0
-500 -500
-1000 -1000
-1500 -1500
-2000 -2000
-0.015 -0.01 -0.005 0 0.005 0.01 0.015 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
Figure ‎5.28 comparison between force-displacement and force-velocity characteristics predicted by

modified Bouc-Wen using different sets of parameters and the measured characteristics under
excitation of 3Hz and 10 mm amplitude at 1 Amp
183
5.6 Summary
A comprehensive experimental testing of hydraulic, hydro-gas and MR dampers has been
performed under different excitation frequencies of between 1 and 11 Hz, and dynamic amplitudes
of up to 12 mm using an electrohydraulic testing machine. The dynamic characteristics of the three
dampers are presented in terms of the force-velocity and force-displacement diagrams. The
damping force dependence on the relative displacement and velocity has been discussed under
different excitation frequencies and with variation of the dynamic amplitudes and excitation current
for the MR damper. Following this, the restoring force surface technique has been used to
represent the dynamic characteristics of the three dampers. This was achieved by fitting the
measured data using Chebyshev polynomials to produce the restoring force surfaces. A validation
between the predicted responses using the Chebyshev polynomial approximation and the
measured data has been presented. The restoring force surface technique offers a good visual
indication of the degree of nonlinearities inherent in the different kinds of dampers. It has been
shown that the MR damper has a wider dynamic range with highly nonlinear dynamic
characteristics than the hydraulic and hydro-gas dampers. Moreover, a modified Bouc-Wen
dynamic parametric model of the MR damper has been presented and its predicted response has
been validated against the measured characteristics of the MR damper. The effect of the model
parameters on the predicted response has been investigated by comparing the predicted
characteristics with the measured dynamic behaviour of the MR damper. Thus, the best set of
parameters to use in the modified Bouc-Wen model to characterise the MR dampers has been
derived.
184
6 CHAPTER 6: DESIGN, MANUFACTURE AND INSTRUMENTATION OF TRACKED
VEHICLE SUSPENSION TEST RIG
6.1 Introduction
This chapter describes the test setup that is used in this study. The hardware and software
components are discussed. First, a description of the seven degrees of freedom of the tracked
vehicle suspension rig is presented. This is followed by an illustration of both the drive mechanism
used for excitation and the sensors used to measure the rig responses. The software and hardware
components include the LabVIEW package and the National Instrument (NI) CompactRio card
which are used together as the data acquisition system and as the controller during all tests. Then,
the experimental procedures are presented. Also, the semi-active control methods and the MR
damper controller are discussed. Finally, the semi-active suspension simulink model of the test rig
is presented.
6.2 Design of test rig

A test rig has been designed to experimentally evaluate the robustness of the semi-active MR
dampers in the tracked vehicle suspension system. The front and side views of the test rig are
shown in Figure 6.1 and Figure 6.2, respectively. The test rig is the physical representation of the
7-DOF tracked vehicle suspension system.
Figure 6.1 Front View of Seven DOF Test Rig (physical model)
185
Rails ‘Hull body’
(sprung mass)
Carriage
‘Road wheel’
(unsprung
mass)
Excitation
masses
Carriages
Figure 6.2 Side View of Seven DOF Test Rig (physical model)
The typical tracked vehicle suspension system consists of four main parts: a sprung mass (hull
body), unsprung masses (road wheels), elastic and damping elements. Springs and dampers are
the elastic and damping elements that connect the hull body with the road wheels. As shown in
Figure 6.3 and Figure 6.3, the aluminium beam of 101.6x101.6x1100 mm represents the hull body.
The beam has two degrees of freedom; bounce in the vertical plane and pitch motion about its
centre of gravity. Five aluminium blocks of 20x101.6x140 mm resemble the unsprung masses
(road wheels). Each wheel has one degree of freedom only (vertical bounce). There are five soft
springs that connect the wheels with the hull body. The rig was designed to accommodate various
numbers of dampers at different locations in the suspension. These different locations offer the
ability to simulate various configurations of the actual tracked vehicles. Five stiffer springs
represents the wheels rigidity and connect the wheels with the drive mechanism.
186
MR damper Soft springs Sprung mass
Unsprung masses Stiffer springs
Figure ‎6.3 Test rig primary components
The use of rails and carriages make the fixing and assembly of the rig easy. The degrees of
freedom of the movable masses are preserved by using the carriages and rails as shown in
Figure 6.4. The carriages are used for securing the vertical bounce of the hull, wheels and
excitation masses (which are used to connect the connecting rod of the drive system to stiffer
springs of the wheels). The carriages slide on five rails that are attached to the main rig frame. The
frame consists of five side beams of 38.1x76.2x850 mm. These side beams are bolted to a base
plate which is anchored to the heavy plate by T-shape studs. The side beams are linked at the top
by a cross beam to prevent the side-to-side motion of the assembly.
The lower part of the test rig is the mechanical drive system which excites the physical model with
a periodic or a swept sine wave excitation. The drive mechanism consists of a steel shaft with five
eccentric discs and five connecting rods. The discs are fixed on the shaft with different orientations.
The phase angle of each disc depends on the wavelength of the road type and on the horizontal
distance between the wheels centre and the hull centre of gravity as described in equation (4.1).
The rotating shaft is supported by six bearings which rest on steel blocks that are bolted to the
base plate as shown in Figure 6.5. Additional description of the est rig components and the
schematic drawings are covered in appendix A.
6.3 Drive system

A geared motor (electric motor coupled with a gear box) along with an inverter are used to drive the
mechanical system at its desired speeds. The whole system can be controlled manually via an
inverter panel or automatically through the PC. The geared motor supplies sufficient torque and
speed to excite the system with the desired excitation. The motor power is 4 kW at 1430 r.p.m. and
the gear box has a 2.17 transmission ratio. The purpose of the gear box is to reduce the speed to a
suitable range that covers the required frequency span. The excitation frequencies ranged from 0
to 10 Hz which corresponds to a motor speed range from 0 to 600 r.p.m.
187
Side beams Cross beam
Rails
Carriage
Figure ‎6.4 Rig main structure with rails and carriages
Connecting rods Bearings Eccentric discs
Figure ‎6.5 Drive mechanism and its components
Coupling Gear box Motor
Figure ‎6.6 Electric geared motor
6.4 Vibration system setup

Figure 6.7 summarizes the data acquisition flow chart used in the experimental tests. The electric
motor controls the speed imposed on the test rig which is excited by the drive mechanism; the
excitation causes the wheel masses and the body to vibrate. Their acceleration and/or velocity are
measured by two accelerometers, z1 and z2 mounted on the body mass and two accelerometers,
zw1 and zr1mounted on the wheel and excitation masses as shown in Figure 6.8. Also, the MR
damper forces, F1 are measured through a Dytran force sensor type 1210C. The collected signals
188
from the rig structure are then amplified and filtered using B&K charge amplifiers model 2635. The
NI-CompactRio card enables the digital signal processing of the analog signals. These signals are
then stored in a personal computer. The CompactRio card generates the required analog output
voltage specified by the control method for the MR dampers. The analog voltage is passed through
the damper power supply unit, which is known as the wonder box, and sent to the MR damper.
The measured vertical accelerations of the body mass are used to calculate the body angular
acceleration based on the kinematics relation between these two measured accelerations. In other
words, if two accelerometers are mounted on the body mass and located at distance L1 and L2 from
the body center of gravity, then their vertical displacements can be expressed in terms of the hull
pitch angle through the following equations:
θ; θ (6.1)
where cg and are the equivalent hull bounce and pitch displacements and L and L2 is the
horizontal distance between the two accelerometers. Double differentiation of the above equations
yields the following accelerations:
θ θ θ θ (6.2)
In the above equations, the pitch angle is assumed to be very small (i.e. ≈ 0 and ≈ ).
Rewriting the equations, the pitch acceleration can be expressed as:
(‎6.3)
θ
CompactRio Card
Body acc/vel
Wheel acc/vel
Excitation acc/vel
MR force
Test Rig
PC
MR wonder box
Figure ‎6.7 Schematic of Test rig and Acquisition instruments
189
Figure ‎6.8 Schematic representation of test rig showing accelerometers mount to measure body
and wheel responses
6.4.1 Data acquisition system and control instrumentation

A NI-CompactRio embedded control and acquisition system is used to acquire all signals as well as
to control the MR dampers. The CompactRio contains three components: a processor running a
real-time operating system (RTOS), field-programmable gate array (FPGA), and I/O modules as
shown inFigure 6.9. The CompactRio system is a LabVIEW software-based data acquisition and
controller card. The processor type is cRio-9022 with 533 MHz, 2 GB storage and 256 MB DRAM.
The reconfigurable chassis type is cRio-9113 with 4 slots which accommodates the input and
output modules. The NI 9215 analog input module with ± 10 V, 100 kS/s and 16-bits is used to
acquire the test rig signals. The NI 9263 analog output module with ± 10 V, 100 kS/s and 16-bits is
used to generate the analog signals to control the MR dampers.
The analog output module produces the necessary voltage to control the MR dampers through the
®
Lord Wonder Box . This control box is a companion product for the MR damper devices. This
control kit requires 12 V DC to produce the necessary current for controlling the damper. The
control box accepts a 0-5 Volt DC from the power supply and generates a variable output of 0-2
Amps depending on the applied voltage.
6.5 Experimental procedures

This section explains the input excitation used throughout the experimental tests. It also discusses
how the collected signals are processed in order to analyze the measured results.
6.5.1 Input excitation

A swept sinusoidal input is chosen to identify the frequency response of the suspension system
with MR dampers. The parameters of the excitation signal for all experimental tests are listed in
Table 6.1 and the time history is shown inFigure 6.10. The excitation frequency range is chosen
based on the natural frequencies of the hull (bounce and pitch which are 3.5 Hz and 4.3 Hz) of the
suspension system. The reader will notice that the excitation amplitude listed in Table 6.1 seems to
190
be small. However, preliminary amplitudes with 5 mm and 10 mm are used but they produce
significant body oscillation.
RTOS FPGA 3AI and 1AO
Figure ‎6.9 CompactRio system architecture; (i) photograph and (ii) schematic representation
10
6
Input excitation (m/s2)
-2
-4
-6
-8
-10
0 100 200 300 400 500 600
Time (s)
Figure ‎6.10 Swept sinusoidal input excitation up to 10 Hz at amplitude of 2.5 mm
Table ‎6.1 Excitation signal parameters

Parameter Value
Starting frequency 0.1 Hz
Target frequency 10 Hz
Time span 600 sec
Increment 0.1 Hz
Amplitude 2.5 mm
191
6.5.2 Signal processing methods
The sampling frequency is chosen to be 500 Hz; this rate is 50 times faster than the Nyquist
frequency which is 10 Hz. The measured data are transferred from the CompactRio memory to the
PC for post processing. The MATLAB environment is used for the analysis of all the measured
data. It is the acceleration and angular acceleration of the hull motion which are chosen for the
suspension criteria. The Fast Fourier Transform (FFT) of the measured time signals is calculated in
order to evaluate the acceleration responses of the suspension. Prior to calculating the FFT, the
acceleration responses are filtered with a third order low pass filter with cut-off frequency of 15 Hz.
6.6 Controllers development

In this study, two systemcontrol strategies, namely: skyhook and hybrid controllers in addition to
one MR damper controller are employed. Each control policy is then designed within the LabVIEW
environment and physically implemented though the use of CompactRio control card.
6.6.1 Skyhook control

The identification of the required signals needed for the control is based on its policy. Recalling the
skyhook control policy from Figure 2.13, it is easier to identify the required signals. The governing
th
equation of the control force for the i MR damper is expressed as follows:
θ θ θ (‎6.4)
θ θ
The controller specifies the desired damping force, based on the body velocity, θ and the
relative velocity across the damper ends, . In the case that these two velocities are in
the same direction and their multiplication is positive, then the maximum damping is produced,
otherwise the damping is minimized. As noticed from the equation, the damping force depends on
a certain gain, and upper end velocity.
6.6.2 Hybrid control

Based on the control policy of the hybrid controller which implies that the hybrid controller is a mix
th
between the Groundhook and skyhook controllers, the governing equation for the i MR damper
can be expressed as follows:
σ σ (‎6.5)
and
θ θ θ
(‎6.6)
θ θ
θ
(‎6.7)
θ
192
In the above equations, the Hybrid controller can be switched to Skyhook or Groundhook
controllers according to the value of variable, this variable takes a value between 0 and 1. For
instance, if the variable equals 1, then the controller is purely Skyhook and if it equals zero, the
controller is Groundhook controller.
6.6.3 MR damper controller

This section presents the Heaviside Step Function (HSF) damper controller. The input or supply
voltage is varied between two values i.e. zero or maximum in order to alter the actual damper
force. The governing equation of the input voltage and the damping force can be written as follows
[34, 51]:
(‎6.8)
where Vmax is the maximum voltage to the current driver of the MR damper, and H(.) is the
Heaviside step function. The current driver for the MR damper can be directly changed but the
generated force by the MR damper cannot be commanded. To force the MR damper to produce
approximately the desired control force , the command voltage vi is selected as follows. When
the MR damper force is equal to the desired force (i.e. ), the voltage applied to the damper
should remain at the present level. If the magnitude of the force produced by the damper is smaller
than the magnitude of the desired control force and the two forces have the same direction, the
voltage applied to the current driver is increased to the maximum level so as to increase the
damper force to closely track the desired force. Otherwise, the commanded voltage is set to zero.
6.7 Semi-active control of test rig model

Figure 6.11 shows the schematic diagram of a semi-active vibration control of the test rig
suspension model with MR dampers. The figure shows the 7 DOF test rig suspension model, the
MR damper model (Modified Bouc-Wen model), the system controller (skyhook or hybrid control)
and the MR damper controller (Heaviside step function). The system controller specifies the
desired damping force, , according to the control policy of the system controller, while the MR
damper controller alters the applied voltage, , to the MR damper in order to make the actual
damping force, track the desired force [34].The objective is to reduce the body vibration caused
by the road excitation, .It should be noted that the test rig can be fitted with one MR damper up to
five MR dampers. For example, when the test rig is equuiped with two MR dampers each damper
needs to be controlled individually. This requires a development of two system controllers in
addition to two MR damper controllers.
193
Figure 6.11 Semi-active control system of test rig suspension with MR damper
The semi-active test rig suspension model with MR dampers is built as a direct representation of
the model differential equations. These equations are derived on the same basis as the passive
suspension model discussed in chapter 3.The equations of the bounce and pitch angle of the body
in addition to the bounce of each road wheel are expressed as follows:
The body bounce:
(6.9)
θ
The body pitch:
θ θ (‎6.10)
The road wheel bounce:
θ- - - , i=1,2,..., 5 (‎6.11)
th
It is noted that represents the damping force of the i MR damper and it is produced by the MR
damper model (Modified Bouc-Wen model). The Simulink model of the semi-active test rig
suspension is built in the same way as the passive suspension model (chapter 3) except for the
controllable damping force of the MR damper that replaces the passive damping force of
hydraulic damper. The model parameters such as the body mass and wheel masses are
measured. In addition, the wheel centre distances are obtained by measuring the horizontal
distances between the body centre of gravity and wheel centres. The suspension and wheel
stiffnesses of the test rig are quoted in Appendix B. Table 6.2 summarizes the test rig suspension
physical parameters used in the simulation of the test rig suspension model.
194
Table ‎6.2 Rig suspension parameters
Description Symbol values Description Symbol values
a st b
Body mass (kg) 63 1 wheel centre (m) 0.396
nd
Body moment of inertia 4.4 2 wheel centre (m) 0.226
2
about lateral axis 'y' (kg.m )
th rd
Equivalent stiffness of i 11,730 3 wheel centre (m) 0.01
torsion bar (N/m)
th th c
Mass of i road wheel (kg) 1 4 wheel centre (m) -0.16
th th
Equivalent stiffness of i 21,730 5 wheel centre (m) -0.284
wheel (N/m)
a: This represents the mass of the Aluminum bar and 35kg additional masses
b: The horizontal distance between the hull centre of gravity and the road wheel centre
c: Negative values denote that the distance is measured behind of the body centre
For the MR damper model, the Modified Bouc-Wen model is used in the simulation. The model
dynamic equations are the same as the model discussed in chapter 5. However, the model
parameters are taken from Table 5.8 (set 4). These parameters gives the best fit between the
predicted model response and the measured MR damper characteristics.
Finally, the system controllers (skyhook and hybrid) and the MR damper controller (Heaviside step
function) are built in Matlab/Simulink environment. The Simulink models which are shown in
Figures 6.12 to 6.14 are developed based on controller equations 6.4 (skyhook control), 6.5 to 6.7
(hybrid control) and 6.8 (Heaviside control).
Figure 6.12 Skyhook controller Simulink model
Figure ‎6.13 Hybrid controller Simulink model
195
Figure 6.14 Heaviside Simulink model of the MR damper controller
6.8 Summary
In this chapter, the test rig which represents a scaled model of the tracked vehicle MR suspension
test rig has been designed and fabricated and its components have been described. This includes
the test rig main parts and the drive system. This is followed by a description of the data acquisition
system and the control devices. Then, the input excitation and the signal processing methods have
been presented. Also, the the system controllers as well as the MR damper control have been
developed and implemented in the Labview environment for the purpose of real time control of the
test rig. Furthermore, the semi-active control of the test rig model has been discussed. This
includes the dscription of the semi-active 7 DOF suspension and its differential equations.
Inaddition, the modified Bouc-Wen MR damper model has been presented. Finally, the simulink
models of the system and MR damper controllers are discuseed.
196
7 CHAPTER 7: SEMI-ACTIVE CONTROL OF TEST RIG SUSPENSION WITH
MAGNETORHEOLOGICAL DAMPERS
7.1 Introduction
This chapter presents a series of experimental tests in order to evaluate the passive and semi-
active suspension rig with MR dampers. The chapter begins with the identification of the primary
system characteristics without any MR dampers. This is achieved by performing a frequency
response function (FRF) test of the undamped suspension under swept sinusoidal excitation. Then,
the dynamic performance of the suspension with passive MR dampers is investigated. Following
the passive suspension analysis, the dynamic performance of the semi-active suspension is
studied. The dynamic behaviour of the semi-active suspension includes the open loop control of
the MR damper followed by the closed loop skyhook and hybrid control analysis using one MR
damper. in addition, semi-active closed loop control using two MR dampers is also, studied. This
chapter ends with a validation of the Simulink model of the undamped and damped suspension
system using skyhook control against the experimental tests. The chapter ends with a summary of
the results and remarks of the performance of the semi-active MR suspension system.
7.2 System identification

This section presents the system identification of the undamped suspension system. Identifying the
undamped suspension is important in describing the resonant frequencies and estimating the
damping ratio. An experimental modal test was performed on the undamped suspension system.
The purpose of this test is to identify the natural frequencies of the suspension and corresponding
damping ratios. Figure 7.1 shows the bounce acceleration transmissibility (which is the ratio of the
vertical body acceleration to the input acceleration) of the undamped suspension under swept
sinusoidal excitation. The suspension is excited with frequency from 0 to 10 Hz as the maximum
motor speed equals to 600 r.p.m. Because of this maximum frequency limit, the wheel resonant
frequencies (about 30 Hz) do not appear in the figure. The figure shows that the maximum
transmissibility occurs at frequencies 3.47 Hz and 4.32 Hz. This implies two natural frequencies of
the body which are the pitch and bounce frequencies respectively. One of the primary aims of the
research is to reduce the transmissibility by using the MR dampers.
12
10
Bounce Acc Transmissibility
0
0 1 2 3 4 5 6 7 8 9 10
Frequency (Hz)
Figure ‎7.1 Bounce acceleration transmissibility of the undamped suspension under swept
sinusoidal excitation
197
The FRF transmissibility is also employed to estimate the damping ratio through the Nyquist plot. In
this plot, the real and imaginary parts are plotted together in the Argand plane. The Nyquist plot
enables the identification of the resonance frequencies and the damping ratio. The Nyquist plot of
the acceleration transmissibility can be obtained by the modal analysis curve fitting according to the
following equation [100]:
(7.1)
where , represents the real part of the transmissibility, , represents the

transmissibility imaginary part and is the loss factor which equals twice the damping ratio. The
above equation is the equation of a circle in the Argand plane and the circle is centred at point
[x,y]=[-1/2,-1/ ] with a radius of . Thus, the loss factor can be obtained by simply
4 2 4
measuring the diameter of the circle of the acceleration transmissibility plot.
Figure 7.2 shows the Nyquist plot of the bounce acceleration transmissibility of the undamped
suspension. The figure shows two modes that correspond to bounce and pitch frequencies (shown
in red colour). In addition, two circles are fitted to the real and imaginary parts (shown in blue) in
order to estimate the damping ratios. The loss factor and associated damping ratios for each mode
are listed in Table 7.1.
Figure ‎7.2 Nyquist plot of Bounce acceleration of undamped suspension system
Table ‎7.1 Estimated loss factor and damping ratios

Response Number of modes Loss factor ( ) amping ratio (ξ)
2 Mode 1 0.0986 0.0493
BA (m/s )
Mode 2 0.1326 0.0663
7.3 Evaluation of suspension performance with passive MR dampers

This section investigates the dynamic behaviour of the suspension system with MR damper (off
state), in other words there is no current applied to the MR damper. The aim is to identify the
optimal MR passive suspension settings that offer the best suspension performance under
sinusoidal excitation. This investigation is twofold: the first is the effect of the locations of MR
198
damper on the suspension dynamic performance and the second is the influence of the number of
MR dampers on the suspension dynamics.
7.3.1 Influence of MR damper locations on suspension performance

The effect of the MR damper locations on the suspension performance is experimentally
investigated under swept sinusoidal excitation. The same suspension configurations used in
chapters 3 and 4 are employd in this investigation. These configurations include S0, S1, S2, S3
and S4 with thirty one alternative locations as shown in Table 3.5. The time histories of the body
acceleration (bounce and pitch) responses of the undamped and damped suspension are
measured under the prescribed excitation. The same mixed objective function (MOF) used in
chaters 3 and 4 is utilized here to evaluate the suspension dynamic performance.It should be noted
that the MOF is calculated using the acceleration and displacement responses of the body.
Therefore, the body displacement and pitch angle are obtained from the time histories of the body
accelerations. This is achievd by double integration of the body acceleration responses and using
the ‘polyfit’ command in Matlab to eliminate the drift in the body displacements.
Table 7.2 shows the calculated MOF values of the undamped suspension and damped suspension
with one MR damper at different locations. Overall, the results show that the MOF values vary
when the damper location is changed. The suspension with MR damper located at wheel stations 1
or 2 has lower MOF values than the undamped suspension. However, the suspension with MR
damper located at wheel stations 3, 4 and 5 has higher MOF values than the undamped
suspension. It is shown that the suspension with MR damper at wheel station 1 has the lowest
MOF value followed by the suspension with MR damper at wheel stations 2 and 5.
Table 7.2 MOF values for suspension with one MR damper

Suspension configurations Damper locations MOF values
S0 - 0.0889
At wh1 0.0706
At wh2 0.0796
S1 At wh3 0.1412
At wh4 0.0990
At wh5 0.0935
Also, the MOF values of the suspension system fitted with two MR dampers are listed in Table 7.3.
the results show the variation of the MOF values according to the damper locations. The
suspension with MR dampers located at wheels 15 has the lowest MOF values. This is followed by
the suspension with MR dampers located at wheels 25 and 23. It should be noted that the
suspension with two MR dampers located at wheels 15 is similar to the damper locations in some
exsisting tracked vehicles. Table 7.4 shows the MOF results of the damped suspension with three
MR dampers at different locations. The suspension with MR dampers located at wheel stations 245
has the lowest MOF values followed by the suspension with MR dampers located at wheel stations
125 and 135. Also, the suspension with MR dampers located at wheel stations 125 is similar to the
damper locations in some actual tracked vehicles. The MOF values of the suspension with four MR
dampers are listed in Table 7.5. it is shown that the suspension with MR dampers located at wheel
199
stations 1245 has the lowest MOF values followed by the suspension with MR dampers located at
wheel stations 2345 and 1345.
It is concluded that the dynamic performance of the tracked vehicle is affected by the damper
locations in the suspension system. The best damper locations which offer better ride
characteristics occur when the dampers are located at the extreme wheels. For example, the
suspension system with two MR dampers located at wheel stations 15 (first and last wheels) offer
the best suspension performance. Moreover, the suspension with three MR dampers located at
wheel stations 125 gives better suspension performance. Furthermore, the suspension with four
MR dampers located at wheel stations 1245 offers the best suspension performance. Overall, the
best suspension configuration obtained from the measured suspension performance match the
suspension configurations obtained from simulation in chapters 3 and 4.
Table ‎7.3 MOF values for suspension with two MR dampers

S0 - 0.0889
At wheel 12 0.1372
At wheel 13 0.1445
At wheel 14 0.1372
At wheel 15 0.1164
At wheel 23 0.1208
S2
At wheel 24 0.1520
At wheel 25 0.1184
At wheel 34 0.1411
At wheel 35 0.1236
At wheel 45 0.1493
Table ‎7.4 MOF values for suspension with three MR dampers

S0 - 0.0889
At wheel 123 0.1744
At wheel 124 0.1839
At wheel 125 0.1514
At wheel 134 0.1749
At wheel 135 0.1533
S3
At wheel 145 0.1583
At wheel 234 0.1688
At wheel 135 0.1838
At wheel 245 0.1331
At wheel 345 0.2415
Table 7.5 MOF values for suspension with four MR dampers

S0 - 0.0889
At Wheel 1234 0.1896
At Wheel 1235 0.1509
S4 At Wheel 1245 0.1393
At Wheel 1345 0.1437
At Wheel 2345 0.1401
7.3.2 Influence of number of MR dampers on suspension performance

Typically, the existing tracked vehicle suspension is equipped with two or three dampers fitted at
the first, second and last wheel locations for vibration improvement. The aim of this section is to
200
experimentally identify the maximum number of dampers that should be used to improve the
suspension performance. For this purpose, the response of the suspension with two, three, four
and five MR dampers is measured. The suspension with dampers fitted to wheels 15, wheels 125,
wheels 1245 and all wheels is employed. These damper locations are selected based on the
suspension with the lowest MOF values which are presented in the previous section. It should be
noted that the suspension with one damper is not included in this analysis because the current
tracked vehicle suspension is equipped with at least two dampers.
Figure 7.3 shows the FFT values of the BA and AA responses of the undamped and damped
suspension system with the selected numbers of MR dampers under swept sinusoidal excitation.
The figure shows that the undamped suspension system has two resonant frequencies which
represent the bounce and angular frequencies. Also, mounting the MR dampers on the suspension
reduce the FFT values of both BA and AA of the suspension response. The suspension with 2 to 5
dampers has a better suspension performance than the undamped suspension. As the damping is
increased by increasing the number of dampers, the response of the damped suspension is
changed. For example, increasing the number of dampers from 3 to 5, the BA response is reduced.
However, the responses of the AA are increased. This confirms that the high damping level is not
always beneficial in the vibration isolation. As can be seen from the figure, the suspension with 3
dampers reduces the peak amplitude while the fully damped suspension does not offer much
improvement. In addition, the AA response of the suspension with 3 dampers is better than the
performance of the suspension with 4 and 5 dampers. Thus, from the cost effective point of view
three dampers are suitable for reducing the suspension vibration. These measured results confirm
not only the results obtained from simulation in chapter 3 but also they match the existing damper
locations in most tracked vehicles suspension system as the best locations.
0.125 0.4
Undamped Undamped
MR at wh15 MR at wh15
MR at wh125 MR at wh125
0.1 MR at wh1245 MR at wh1245
MR at all wheels 0.3 MR at all wheels
Bounce Acc (m/s2)
0.075
0.2
0.05
0.1
0.025
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Frequency (Hz) Frequency (Hz)
Figure ‎7.3 Bounce and angular acceleration responses of suspension with different numbers of
passive MR dampers at [wh15, wh125 and all wheels] under swept sinusoidal excitation
7.4 Semi-active control of test rig with one MR damper

In order to show the effectiveness of the MR dampers in vibration attenuation, the suspension with
semi-active control is evaluated. The subsequent sections investigate the dynamics of the semi-
active suspension system with MR dampers using passive, skyhook and hybrid control methods.
201
7.4.1 Open Loop control of test rig using one MR damper
This section evaluates the effect of the open loop control on the behaviour suspension system
incorporating one MR damper. In the open loop control states, five different currents have been
chosen to represent the open loop cases. The damper is energized to each constant current and
the MR damper behaves like a regular passive damper. The five currents represent the range from
0 Amp to 0.5 Amp with an increment of 0.1 Amp. The selected current values require 0 to 5 volts
DC to be applied to the MR power supply. The maximum damper current is about 2 Amp but it is
found that a current up to 0.5 Amp offers acceptable damping force for the tested structure.
Figure 7.4 shows the FFT response of the BA and AA of the undamped and damped suspension
system with MR damper located at wheel 4. Both the BA and AA responses for suspension with the
MR damper in the off state (0 Amp) are better than the undamped suspension response. Overall,
the results show that the damper with open loop passive (MR off) control minimizes the peak
amplitude of the BA and AA responses. However, it is seen that the damper in the MR off state (0
Amp) causes an increase in the system stiffness such that the system resonance frequencies
increase in value. At a current of 0.1 Amp, the overall response is slightly greater than the
response at 0 Amp (MR off). At currents of 0.2 Amp to 0.5 Amp, the overall BA and AA responses
are considerably greater than the corresponding responses at 0 Amp (MR off). Also, there is
considerably shift in the pitch and bounce frequencies. This is due to the stiffening of the MR
damper at currents above 0.1 Amp. This stiffening characteristic is due to the yield and formation of
plug flow in the MR fluid at these current levels.
0.12 0.7
Undamped Undamped
MR off (0 Amp) MR off (0 Amp)
0.1 Amp 0.6 0.1 Amp
0.1
0.2 Amp 0.2 Amp
0.5
0.08
Bounce Acc (m/s2)
0.4
0.06
0.3
0.04
0.2
0.02
0.1
0 0
0 2 4 6 8 0 2 4 6 8
0.14 Frequency (Hz) 1.4 Frequency (Hz)
0.1 Amp 0.1 Amp
0.2 Amp 0.2 Amp
0.12 0.3 Amp 1.2 0.3 Amp
0.4 Amp 0.4 Amp
0.1 0.5 Amp 1 0.5 Amp
Bounce Acc (m/s2)
0.08 0.8
0.06 0.6
0.04 0.4
0.02 0.2
0 0
0 2 4 6 8 0 2 4 6 8
Figure ‎7.4 Bounce and angular acceleration for different damper currents
202
The results of the open loop control of the MR suspension reveal that increasing the damping is
helpful to minimize the suspension vibration. However, this method provides constant currents to
the MR damper which at certain current values increases not only the damping but also the system
stiffness significantly and, therefore, reduces the suspension performance. Therefore, the semi-
active closed loop control of the MR damper, which continuously changes the current, offers the
optimal MR behaviour.
7.4.2 Skyhook closed loop control of test rig using one MR damper
In this section, the suspension dynamic performance with closed loop skyhook control is
investigated. The MR damper is mounted at wheel station number four. As discussed in
section 6.6.1, the skyhook controller continuously specifies the desired force based on the control
policy. As a result, the damper controller using the MR power supply varies the output current to
the MR damper. Thus, the damper generates a damping force that closely tracks the desired force.
Figure 7.5 illustrates the execution of the skyhook control as the damper follows the control law.
The time history of the velocity of the body (v1) at the upper end of the damper is plotted along with
the time history of the relative velocity (v12) across the damper ends and the command voltage time
history, V[0, 5] volts. These time histories represent data samples measured around the body
natural frequencies of the suspension response under swept sinusoidal excitation. The results
reveal that the voltage applied to the MR damper varies between zero and 5 volts according to the
sign of the product of the two velocities (the details are given in section 6.6.3). When the product of
v1 and v12 is positive, the controller generates maximum force and consequently the damper
controller increases the voltage of the MR damper so as to increase the damper force to closely
match the controller force. Otherwise, the voltage is zero volts offering minimum damping force.
It can be seen that the semi-active closed loop control continuously switches between the
maximum and the minimum damping levels. Hence, the MR damper can be controlled to generate
a damping force that adapts to the suspension uncertainties. Conversely, in the open loop control
mode, the damping force is always set at the maximum level all the time preventing the damper
from varying its damping force.
0.15
v1
v12
0.1 voltage*10-2
v1(m/s),v12 (m/s),voltage (volt)
0.05
-0.05
-0.1
-0.15
200 200.2 200.4 200.6 200.8 201
Time (Hz)
Figure ‎7.5 Sample time traces of skyhook control method execution
203
The effect of the skyhook closed loop control gains on the continuous semi-active suspension
system is studied. The control gains are adjusted to change the dynamics of the suspension
system with MR dampers. The gain, constant parameter ‘Gs’ in equation (6.4) varies discretely from
0.1 to 500. The gain value determines the damping (current) levels between the two boundaries
(minimum and maximum damping). Figure 7.6 shows the results of the bounce and pitch
acceleration responses of the suspension system with continuous skyhook control. As the gains
are increased, the peak amplitude of the BA is reduced around the first resonant frequency of the
body. This improves the suspension performance and consequently offers smooth ride dynamic
characteristics. At relatively high frequency over 5 Hz, the peak amplitude of the BA is increased
when the gain is increased indicating that the skyhook controller is very sensitive to gain variation.
The AA response shows that increasing the controller gain reduces the peak amplitude and
improves the suspension ride behaviour. However, there is a slight increase in the AA peak
amplitude when the controller gain is increased over 200 Ns/m.
0.06
Skyhook [G=0.1]
Skyhook [G=0.5]
Skyhook [G=10]
0.05
Skyhook [G=50]
Skyhook [G=100]
Skyhook [G=200]
0.04 Skyhook [G=500]
Bounce Acc (m/s2)
0.03
0.02
0.01
0
2 3 4 5 6 7 8
Frequency (Hz)
0.4
Skyhook [G=0.1]
Skyhook [G=0.5]
0.35 Skyhook [G=10]
Skyhook [G=50]
0.3 Skyhook [G=100]
Skyhook [G=200]
Skyhook [G=500]
0.25
0.2
0.15
0.1
0.05
0
2 3 4 5 6 7 8
Frequency (Hz)
Figure ‎7.6 Effect of gains of the skyhook control on BA and AA responses of the suspension with
MR damper under swept sinusoidal excitation
As expected, the skyhook control affects the body accelerations but at the expense of increasing
the wheel acceleration responses. Figure 7.7 shows the effect of skyhook gains on the wheel
acceleration responses. The results show that increasing the controller gain from 0.1 to 50 reduces
204
the wheel acceleration amplitude. However, when the gain is further increased, the wheel
acceleration amplitude is increased.
0.045
Skyhook [G=0.1]
Skyhook [G=0.5]
0.04
Skyhook [G=10]
Skyhook [G=50]
0.035 Skyhook [G=100]
Skyhook [G=200]
Wheel Bounce Acc (m/s2)
0.03 Skyhook [G=500]
0.025
0.02
0.015
0.01
0.005
0
2 3 4 5 6 7 8
Frequency (Hz)
Figure 7.7 Effect of gains of the skyhook control on wheel acceleration response of the suspension
with MR damper under swept sinusoidal excitation
7.4.3 Hybrid closed loop control of test rig using one MR damper
As mentioned in section 6.6.2, the hybrid closed loop control affects the sprung and unsprung
mass accelerations. Two main factors, which are the controller gain and the weighting factor ,
influence the hybrid control performance. Therefore, the effect of varying the controller gain and
weighting factors on the suspension performance are studied. The evaluation of the hybrid control
is carried out by holding the weighting factor constant at 0.5 while changing the hybrid gain from
100 to 1000 Ns/m. In order to examine the effect of varying the weighting factor on the hybrid
controller, the controller gain of 1000 Ns/m is held constant while the weighting factor is varied from
0 to 1 with increment 0.25.
Figure 7.8 show the responses of the BA, AA and wheel acceleration of the suspension under
swept sinusoidal excitation for equals 0.5 and different gain values. This specific hybrid
configuration represents equal contribution from the skyhook and groundhook control. In other
words, the common trade-off in skyhook and groundhook is eliminated in this configuration. This
trade-off comes from the difficulty to simultaneously improve both the responses of the sprung and
unsprung masses. As the hybrid gain increases from 500 Ns/m to 1000 Ns/m, the bounce and
angular acceleration of the body are attenuated. In addition, the wheel bounce acceleration is
improved when the hybrid gain is increased. This shows that the hybrid control eliminates the
trade-off in the suspension performance.
The effect of varying while holding the hybrid gain constant at 1000 Ns/m is shown in Figure 7.9.
The results reveal that when equals , the controller switched to pure skyhook and the body
acceleration responses are attenuated at the expense of increasing the wheel bounce acceleration.
onversely, when equals ero, the wheel bounce acceleration is improved as the hybrid controller
is changed to groundhook control.
205
It can be concluded that the hybrid control enhances the overall suspension performance.
However, determination of the appropriate gain and weighting factor is still a challenge in improving
the hybrid control performance.
0.05
MR off
Hybrid [G=100]
Hybrid [G=200]
Hybrid [G=500]
0.04 Hybrid [G=1000]
Bounce Acc (m/s2)
0.03
0.02
0.01
0
2 3 4 5 6 7 8
Frequency (Hz)
0.35
MR off
Hybrid [G=100]
Hybrid [G=200]
0.3
Hybrid [G=500]
Hybrid [G=1000]
0.25
0.2
0.15
0.1
0.05
0
2 3 4 5 6 7 8
Frequency (Hz)
0.04
Hybrid [G=100]
Hybrid [G=200]
0.035 Hybrid [G=500]
Hybrid [G=1000]
0.03
0.025
0.02
0.015
0.01
0.005
0
2 3 4 5 6 7 8
Frequency (Hz)
Figure ‎7.8 Effect of gains of the hybrid control on BA, AA and wheel responses of the suspension
with MR damper under swept sinusoidal excitation (tau=0.5)
206
0.05
MR off
Hybrid [tau=0]
Hybrid [tau=0.25]
Hybrid [tau=0.5]
0.04 Hybrid [tau=0.75]
Hybrid [tau=1]
Bounce Acc (m/s2) 0.03
0.02
0.01
0
2 3 4 5 6 7 8
Frequency (Hz)
0.5
MR off
Hybrid [tau=0]
Hybrid [tau=0.25]
Hybrid [tau=0.75]
Hybrid [tau=1]
0.3
0.2
0.1
0
2 3 4 5 6 7 8
Frequency (Hz)
0.06
Hybrid [tau=0]
Hybrid [tau=0.25]
Hybrid [tau=0.5]
Hybrid [tau=1]
0.04
0.03
0.02
0.01
0
2 3 4 5 6 7 8
Frequency (Hz)
Figure ‎7.9 Effect of weighting factor, of the hybrid control on BA, AA and wheel responses of the
suspension with MR damper under swept sinusoidal excitation (G=1000 Ns/m)
207
7.5 Semi-active closed loop control of test rig using two MR dampers
Additional semi-active vibration control studies of the MR suspension rig equipped with two MR
dampers located at wheel stations 1 and 5 are used to show the effectiveness of the MR dampers
in vibration isolation. Two control methods, which are the skyhook and hybrid controls, are used as
system controllers in this investigation. In addition, the Heaviside step function is used as the MR
damper controller. It should be noted that the semi-active control of the test rig using two MR
dampers is performed by individually controlling the performance of each damper. For this reason,
two system controllers as well as two MR damper controllers are developed in the LabVIEW
environment. The National Instrument’s CompactRio card is used as an interface between the rig
hardware and developed control algorithms. The influence of the skyhook and hybrid control gains
is also investigated in this investigation. The suspension system is excited by swept sinusoidal
excitation of frequencies from 0 to 10 Hz and dynamic amplitude of 2.5 mm. It should be noted that
in the first instance, the responses of the undamped suspension and the damped suspension with
two MR dampers (off state) are measured. Then, the responses of the controlled suspension
system (using skyhook and hybrid control methods) are measured.
Figure 7.10 shows the bounce and angular accelerations of the vehicle body under swept
sinusoidal excitations. The response of the undamped suspension is compared to the responses of
the suspension with MR passive state as well as the controlled suspension using skyhook control
with different gains of 50 and 100 Ns/m. It is shown that the passive and controlled MR suspension
have better suspension performance than the performance of the undamped suspension system.
In addition, the performance of the controlled suspension using skyhook control with gain 50 Ns/m
offers slight improvement in the bounce acceleration response than the performance of the
suspension with MR passive. However, the performance of angular acceleration of the suspension
with MR passive is better than the performance of the controlled suspension using skyhook control.
Also, the figure shows that when the skyhook control gain is increased to 100 Ns/m both the
bounce and angular acceleration responses are significantly improved and the suspension
performance becomes much better than the performance of the suspension with MR passive. This
confirms the results obtained from the experimental tests in section 7.4.2.
0.12 0.4
Undamped Undamped
MR passive MR passive
0.1 Skyhook [G=50] Skyhook [G=50]
Skyhook [G=100] Skyhook [G=100]
0.3
Bounce Acc (m/s2)
0.08
0.06 0.2
0.04
0.1
0.02
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Figure ‎7.10 Skyhook control of BA and AA of the suspension with two MR dampers located at
wheels 1 and 5
208
Similarly, Figure 7.11 shows the measured performance of the undamped, MR passive and
controlled suspension with hybrid control under the same excitation. The controlled MR suspension
system is evaluated using hybrid control with two different gains which are 10 and 50 Ns/m at a
constant weighting factor of 0.5. The value of the weighting factor is chosen to equally balance the
hybrid controller between the skyhook and groundhook controls. The figure shows that the
measured performance of the MR passive and controlled suspension is better than the
performance of the undamped suspension system. In addition, it can be seen that the controller
gain also affects the performance of the suspension system with two MR dampers. For example,
the controlled suspension with the hybrid control gain of 10 Ns/m has no significant improvement
compared with the performance of the suspension with MR passive. However, the use of hybrid
control gain of 50 Ns/m considerably improves the performance of the suspension compared with
the performance of the suspension with MR passive.
It can be concluded that the semi-active control of the test rig with two MR dampers show the
effectiveness of the MR dampers in vibration isolation of the multiple suspension system. The
controlled characteristics of the MR dampers enable the suspension to adapt to the change in the
vehicle dynamics and road excitations.
0.12 0.4
Undamped Undamped
MR passive MR passive
0.1 Hybrid [G=10, tau=0.5] Hybrid [G=10, tau=0.5]
Hybrid [G=50, tau=0.5] Hybrid [G=50, tau=0.5]
0.3
Bounce Acc (m/s2)
0.08
0.06 0.2
0.04
0.1
0.02
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Figure 7.11 Hybrid control of BA and AA of the suspension with two MR dampers located at wheels
1 and 5
7.6 Comparisons of open loop and closed loop semi-active control of test rig using one
and two MR dampers
In this section, the performance of the suspensions with one and two MR dampers in the open loop
passive state (0 Amp) and active states (0.1 to 0.5 Amp) with the closed loop skyhook and hybrid
controller actions are compared. For the sake of brevity, no new figures are presented. Rather, the
figures that have already been presented in sections 7.4 and 7.5 are referenced for the
comparisons.
The results of the open loop control, skyhook and hybrid closed loop control of the test rig using
one MR damper (MRD) located at wheel position 4 are presented in Figures 7.4 (passive), 7.6 - 7.7
(skyhook), and 7.8 - 7.9 (hybrid). For open loop control, Figure 7.4 shows that the best
performance is achieved by using the one damper in its passive off state (0 Amp). For skyhook
semi-active control, Figures 7.6 and 7.7 show that the best performance is achieved by using the
209
one MR damper in its active state with skyhook control and with skyhook controller gain of G = 50
Ns/m. Also, for hybrid semi-active control, Figures 7.8 and 7.9 show that the best performance is
achieved by using the one MR damper in its active state with hybrid control and with hybrid
controller gain of G = 1000 Ns/m and weighting factor of  = 0.5.
The results of the open loop control, skyhook control and hybrid control of the test rig using two MR
dampers located at wheel positions 1 and 5 are presented in Figures 7.10 and 7.11. The figures
show that the MR passive state gives a better suspension performance than the undamped
suspension. However, the two passive MR dampers cause a significant increase in the system
stiffnesses such that the pitch and bounce frequencies are greatly increased. The figures also
show that the best suspension performances is achieved by using the MR damper and skyhook
control with controller gain of G = 100 Ns/m, or by using the MR damper and hybrid control with
controller gain of G = 50 Ns/m and weighting factor of  = 0.5. By examining Figure 7.12 and Figure
7.13 further, it can be deduced that the hybrid controller gives a slightly better performance than
the skyhook controller. Therefore, it can be concluded that for the suspension with two MR
dampers, the hybrid controller gives the best performance.
Also, comparing Figures 7.4 with Figure 7.10, it can be seen that the suspension system with two
passive MR dampers located at wheels 1 and 5 gives a better performance than the suspension
with one passive MR damper located at wheel 4. Comparing Figures 7.6 - 7.7 (1 MRD, skyhook)
with Figure 7.10 (2 MRDs, skyhook) and Figures 7.8 - 7.9 (1 MRD, hybrid) with Figure 7.11 (2
MRDs, hybrid), it is seen that the use of two semi-active MR dampers located at wheels 1 and 5
enable a higher degree of semi-active control than the use of only one semi-active MR damper
located at w4. Overall, it can be concluded that the suspension system is improved by using two
passive MR dampers than one MR damper, and it is much more improved by using the hybrid
controller with the two MR dampers.
7.7 Validation of test rig model

This section presents the validation of the measured performance of the undamped and controlled
MR suspension against the computer model. First, the measured and predicted responses of the
body accelerations of the undamped suspension are presented. Then, the predicted performance
of the suspension Simulink model using skyhook and hybrid control of one MR damper is
compared to the measured test rig responses under sinusoidal excitation of 3.5 Hz.
7.7.1 Frictional effects in test rig

This section presents a description of the frictional effect in the test rig moving parts. Recalling the
test rig components described in chapter 6, the drive mechanism (shaft and connecting rods)
produces reciprocating forces to excite the test rig. These forces are transmitted to the road wheel
masses though the stiffer springs (which represent the wheel stiffnesses). The vertical bounce of
the wheel masses is transferred to the vehicle body which is free to move in the vertical plane.
Although, the excitation masses are attached to low friction carriages (five carriages), some
resistance appears from the side forces of the crank shaft. These forces prevents the carriages
210
from moving freely on the rails. Consequently, the carriages experience a high level of frictional
resistances on the rails.
It should be noted that the friction effect was unknown and, therefore, it was not initially introduced
in the Simulink model of the test rig. It was introduced later by calibrating the predicted and
measured data to account for this frictional resistance. This was achieved by comparing the input
acceleration and bounce acceleration at wheel 1 of the measured and predicted responses. For
this reason, the undamped suspension of the test rig was excited with a swept sinusoidal excitation
of up to 10 Hz at 2.5 mm amplitude. In addition, the same response of the Simulink model was
generated using the same excitation (chirp signal).
Figure ‎7.12 shows the comparison between the predicted and measured responses of the
undamped suspension system. It is seen that the predicted input acceleration is close to the
measured acceleration at wheel one. This confirms that both the test rig and the Simulink model
generate the same excitation. Although the predicted acceleration response at wheel 1 shows good
match with the measured acceleration in terms of the dominant peaks, the measured response has
lower amplitudes. This reduction in the acceleration amplitude is due to the friction inherent in the
test rig’s moving masses which is not included in the Simulink model. So, it is expected that the
predicted body responses should have the same behaviour.
In order to include this frictional effect in the computer model, a certain amount of damping is
introduced in the predicted response of the undamped suspension in order to make it to correlate
closer to the measured responses.
0.15 0.15
Predicted Predicted
Measured Measured
0.125 0.125
wheel1 Acc. (m/s2)
0.1 0.1
Input Acc. (m/s2)
0.075 0.075
0.05 0.05
0.025 0.025
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Figure ‎7.12 predicted and measured input acceleration and wheel 1 acceleration of the undamped
suspension under swept sinusoidal excitation and amplitude of 2.5 mm
7.7.2 Validation of undamped suspension natural frequencies

The prediction of the Simulink model is validated by the measured data from the test rig. The
validation includes two sets of data which are the resonant frequencies of the suspension and the
acceleration responses due to fixed sinusoidal and swept sinusoidal excitations.
The first set of validation includes the comparison between the model and the rig natural
frequencies. The natural frequencies and the mode types of the undamped suspension model are
predicted based on the solution of the undamped free vibration of the suspension model just like
211
the procedures discussed in section 3.5. The test rig model parameters used are taken from
Table 6.2. It should be noted that the values quoted in this table for the compression springs are
modified because these springs are attached to the test rig through spring pillows which increase
the spring stiffness. For the suspension spring, the stiffness is increased by a factor of 1.071 that
corresponds to a reduction in the spring free length by one coil. Similarly, the stiffness of the wheel
springs is increased by 1.261 that corresponds to a reduction in the spring free length by two coils.
The predicted natural frequencies of the undamped suspension are listed in Table 7.6. The results
show that the model has two natural frequencies corresponding to the bounce and pitch motions of
the body and five bounce resonant frequencies for the road wheels.
The predicted natural frequencies (body bounce and pitch) of the undamped suspension model are
compared with the ones measured from the rig. It is noted that the measured natural frequencies
are estimated from the frequency response function (FRF) of the hull motions as presented in
section 7.2. As shown in Table 7.7, the measured and predicted frequencies show a good
agreement between the proposed model and the test rig.
Table ‎7.6 Predicted undamped natural frequencies and associated modes

Mode number Undamped natural frequency (Hz) Mode type
1 3.71 Hull pitch
2 4.35 Hull bounce
3 31.82 Wheel #1 bounce
Table ‎7.7 Measured and predicted hull natural frequencies

Mode Undamped natural frequency (Hz) M -P
Mode type Measured Values Predicted Values Error (%) = x
number M
(MV) (PV)
1 Hull pitch 3.5 3.71 6
2 Hull bounce 4.35 4.33 0.45
7.7.3 Validation of undamped suspension acceleration responses

The second set of validation consists of the comparison between the predicted and measured
acceleration responses of the damperless suspension (i.e. without any damper). It should be noted
that the damperless rig still possessed some level of damping due to friction, joints, air damping
and internal damping of the helical coil springs as discussed earlier. This background damping of
the rig was introduced in the Simulink model as follows: (i) using viscous damping of constant
damping coefficients 0, 10, 40, 60 and 80 Ns/m between the 5 wheel masses and the body mass,
and (ii) using a viscous damper with constant damping coefficient of 360 Ns/m at wheel station
number 3. The damperless suspension rig is subjected to: (i) fixed sinusoidal excitations at
frequencies of 2, 3.5, 4.3 and 6 Hz, and (ii) swept sinusoidal excitation from 0 to 10 Hz. Then, the
bounce and angular accelerations of the body are measured under swept sinusoidal excitation
only. The simulation is performed using the same type of swept sinusoidal excitation and amplitude
for duration of 600 seconds and a sufficient sampling frequency of 1 kHz. In addition, the simulation
212
is performed for fixed sinusoidal excitations. Then, the predicted hull bounce and pitch acceleration
responses are obtained.
(1) Responses under fixed sinusoidal excitations
In this case, viscous damping of constant damping coefficients 0, 10, 40, 60 and 80 Ns/m is
assumed to exist between the 5 wheel masses and the body mass. These represent the frictional
damping between the carriages, which support the 5 wheel masses and body mass on the 5 slides.
At excitation frequency of 2 Hz, Figure 7.13 shows the response when there is no damping, while
Figures 7.14 to 7.17 show the responses when the damping coefficients are 10, 40, 60 and 80
Ns/m, respectively. It is seen that when there is little or no damping, the resonance frequencies at
3.5 and 4.3 Hz are excited although the excitation is at the lower frequency of 2 Hz. But when the
damping level is increased, the responses at the resonance frequencies become suppressed.
Similarly, Figures 7.18 to 7.22 and Figures 7.23 to 7.27 show the corresponding responses at
excitation frequencies of 3.5 Hz and 4.3 Hz, which are the resonance frequencies. The results
show that the response is primarily at the excitation frequency of 3.5 Hz or 4.5 Hz. The response at
the resonance frequency that is different from the excitation frequency is very little when there is
little or no damping present. When the damping is increased, there is hardly any response at the
resonance frequency that is different from the excitation frequency. But when the excitation
frequency is increased to 6 Hz, which is greater than the two resonance frequencies, Figures 7.28
to 7.32 show that responses are excited at the resonance frequencies when there is little or no
damping present. When the damping level is increased, the responses at the resonance
frequencies become suppressed.
These results show that damping plays a significant role in the predicted responses. In the
experimental rig, there is frictional damping present between the carriages of the wheel masses
and the slides on which they move as well as between the carriage of the body mass (the beam)
and the slide on which it moves. This frictional damping needs to be taken into account in the
simulations. The approach adopted to do this is described in the next sub-section on “Responses
under swept sinusoidal excitations”.
213
2 0.8
C=0 Ns/m
1.5 0.7
1 0.6
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.5 0.5
0 0.4
-0.5 0.3
-1 0.2
-1.5 0.1
-2 0
0 1 2 3 4 5 0 2 4 6 8 10
C=0 Ns/m
4
10
3.5

5 3
2.5
0
2
-5 1.5
1
-10
0.5
-15 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure ‎7.13 Predicted acceleration responses of undamped suspension under sinusoidal excitation
of 2 Hz and damping coefficient (C=0 Ns/m)
1.5 0.4
C=10 Ns/m
0.35
1
0.3
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.5
0.25
0 0.2
0.15
-0.5
0.1
-1
0.05
-1.5 0
0 1 2 3 4 5 0 2 4 6 8 10
8 Time (s) 2 Frequency (Hz)
C=10 Ns/m
6 1.8
1.6
4
1.4
2
1.2
0 1
0.8
-2
0.6
-4
0.4
-6 0.2
-8 0
0 1 2 3 4 5 0 2 4 6 8 10
214
1 0.14
C=40 Ns/m
0.8
0.12
0.6
0.4 0.1
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.2
0.08
0
0.06
-0.2
-0.4 0.04
-0.6
0.02
-0.8
-1 0
0 1 2 3 4 5 0 2 4 6 8 10
C=40 Ns/m
1.6
6
1.4
4

1.2
2 1
0 0.8
0.6
-2
0.4
-4
0.2
-6 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure 7.15 Predicted acceleration responses of undamped suspension under sinusoidal excitation
0.8 0.14
C=60 Ns/m
0.6 0.12
0.4
0.1
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
0.08
0
0.06
-0.2
0.04
-0.4
-0.6 0.02
-0.8 0
0 1 2 3 4 5 0 2 4 6 8 10
Time (s) 1.8 Frequency (Hz)
8
C=60 Ns/m
1.6
6
1.4
4
1.2
2 1
0 0.8
0.6
-2
0.4
-4
0.2
-6 0
0 1 2 3 4 5 0 2 4 6 8 10
215
0.8 0.12
C=80 Ns/m
0.6
0.1
0.4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.08
0.2
0 0.06
-0.2
0.04
-0.4
0.02
-0.6
-0.8 0
0 1 2 3 4 5 0 2 4 6 8 10
C=80 Ns/m
1.6
6
1.4
4

1.2
2 1
0 0.8
0.6
-2
0.4
-4
0.2
-6 0
0 1 2 3 4 5 0 2 4 6 8 10
30 14
C=0 Ns/m
12
20
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)
10
8
0
6
-10
4
-20
2
-30 0
0 1 2 3 4 5 0 2 4 6 8 10
C=0 Ns/m
60
100
50
50
40
0
30
-50
20
-100
10
-150 0
0 1 2 3 4 5 0 2 4 6 8 10
of 3.5 Hz and damping coefficient (C=0 Ns/m)
216
20 10
C=10 Ns/m
15 9
8
10
7
Bounce Acc (m/s2)
Bounce Acc (m/s2)

5
6
0 5
4
-5
3
-10
2
-15 1
-20 0
0 1 2 3 4 5 0 2 4 6 8 10
C=10 Ns/m
80
50
60
40

40
20
0 30
-20
20
-40
-60
10
-80
-100 0
0 1 2 3 4 5 0 2 4 6 8 10
8 5
C=40 Ns/m
6 4.5
4
4
3.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
3
0 2.5
2
-2
1.5
-4
1
-6 0.5
-8 0
0 1 2 3 4 5 0 2 4 6 8 10
C=40 Ns/m
40
30
30
20 25
10
20
0
15
-10
-20 10
-30
5
-40
-50 0
0 1 2 3 4 5 0 2 4 6 8 10
217
5 3.5
C=60 Ns/m
4
3
3
2 2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

1
2
0
1.5
-1
-2 1
-3
0.5
-4
-5 0
0 1 2 3 4 5 0 2 4 6 8 10
C=60 Ns/m
30
20
20

10
15
10
-10
-20
5
-30
-40 0
0 1 2 3 4 5 0 2 4 6 8 10
4 3
C=80 Ns/m
3
2.5
2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
2
1
0 1.5
-1
1
-2
0.5
-3
-4 0
0 1 2 3 4 5 0 2 4 6 8 10
C=80 Ns/m
20
15 20
10
5 15
-5 10
-10
-15 5
-20
-25 0
0 1 2 3 4 5 0 2 4 6 8 10
218
15 6
C=0 Ns/m
5
10
Bounce Acc (m/s2)
Bounce Acc (m/s2)

4
5
0
2
-5
1
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
C=0 Ns/m
40
30 20
20

10 15
-10 10
-20
-30 5
-40
-50 0
0 1 2 3 4 5 0 2 4 6 8 10
10 4.5
C=10 Ns/m
8 4
6 3.5
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
3
2
2.5
0
2
-2
1.5
-4
-6 1
-8 0.5
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
C=10 Ns/m
30
20
20
10
15
10
-10
-20
5
-30
-40 0
0 1 2 3 4 5 0 2 4 6 8 10
219
6 4
C=40 Ns/m
3.5
4
3
Bounce Acc (m/s2)
Bounce Acc (m/s2)

2
2.5
0 2
1.5
-2
1
-4
0.5
-6 0
0 1 2 3 4 5 0 2 4 6 8 10
C=40 Ns/m
20
20

10
15
10
-10
5
-20
-30 0
0 1 2 3 4 5 0 2 4 6 8 10
5 3.5
C=60 Ns/m
4
3
3
2 2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
1
2
0
1.5
-1
-2 1
-3
0.5
-4
-5 0
0 1 2 3 4 5 0 2 4 6 8 10
C=60 Ns/m
20
15 20
10
5 15
-5 10
-10
-15 5
-20
-25 0
0 1 2 3 4 5 0 2 4 6 8 10
220
4 3.5
C=80 Ns/m
3 3
2
2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

1
2
0
1.5
-1
1
-2
-3 0.5
-4 0
0 1 2 3 4 5 0 2 4 6 8 10
C=80 Ns/m
20 18
15 16
10 14

5 12
0 10
-5 8
-10 6
-15 4
-20 2
-25 0
0 1 2 3 4 5 0 2 4 6 8 10
1.5 0.5
C=0 Ns/m
0.45
1
0.4
0.5 0.35
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.3
0
0.25
-0.5
0.2
-1 0.15
0.1
-1.5
0.05
-2 0
0 1 2 3 4 5 0 2 4 6 8 10
C=0 Ns/m
6
10
5
5
4
0
3
-5
2
-10
1
-15 0
0 1 2 3 4 5 0 2 4 6 8 10
221
1.5 0.5
C=10 Ns/m
0.45
1 0.4
0.35
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.5 0.3
0.25
0 0.2
0.15
-0.5 0.1
0.05
-1 0
0 1 2 3 4 5 0 2 4 6 8 10
C=10 Ns/m
6
10
5

5
4
3
0
2
-5
1
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
1 0.7
C=40 Ns/m
0.8
0.6
0.6
0.4 0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.2
0.4
0
0.3
-0.2
-0.4 0.2
-0.6
0.1
-0.8
-1 0
0 1 2 3 4 5 0 2 4 6 8 10
C=40 Ns/m
8
6
6
4 5
2
4
0
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
222
1 0.7
C=60 Ns/m
0.8
0.6
0.6
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.4
0.2 0.4
0 0.3
-0.2
0.2
-0.4
0.1
-0.6
-0.8 0
0 1 2 3 4 5 0 2 4 6 8 10
C=60 Ns/m
8
6
6
4 5

2
4
0
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
1 0.7
C=80 Ns/m
0.8
0.6
0.6
0.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.4
0.2 0.4
0 0.3
-0.2
0.2
-0.4
0.1
-0.6
-0.8 0
0 1 2 3 4 5 0 2 4 6 8 10
C=80 Ns/m
8
6
6
4 5
2
4
0
3
-2
-4 2
-6
1
-8
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
223
(2) Responses under swept sinusoidal excitations
In this case, a viscous damper with constant damping coefficient of 360 Ns/m is assumed at wheel
station number 3. That is, the damper is assumed to exist between wheel mass 3 and the body
mass. This assumption is made as the body mass is supported by a carriage on the same slide as
wheel mass 3. Also, this assumes that the effects of the frictional damping in all the slides can be
concentrated into a fictitious damper between wheel mass 3 and the body mass.
Figure 7.33 shows the absolute values of the FFT responses of the measured and predicted
accelerations. The overall comparison shows good agreement between the measured and
predicted accelerations in terms of the peak amplitudes. However, there is a slight difference in the
predicted pitch natural frequency. This difference is most probably due to the errors associated with
the spring stiffness and the large oscillations observed in the experimental tests. It can be
concluded that these results confirm that there is a close match between the undamped responses
of the theoretical model and the test rig. This assumption that all the frictional damping effects can
be represented by viscous damping between wheel mass 3 and the body mass is used for all
subsequent simulations.
0.12 0.35
Predicted Predicted
Measured Measured
0.3
0.1
0.25
Angular Acc. (rad/s2)
Bounce Acc. (m/s2)
0.08
0.2
0.06
0.15
0.04
0.1
0.02
0.05
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Figure ‎7.33 Predicted and measured body acceleration responses of undamped suspension under
swept sinusoidal excitation and amplitude of 2.5 mm
7.7.4 Validation of damped suspension with MR passive damper

In this section, the predicted and the measured test rig responses of the suspension using one MR
damper in a passive state are compared. The damped suspension system is equipped with one
MR damper at five alternative locations as described in section 7.3.1. The measured and predicted
responses of the damped suspension system are due to the swept sinusoidal excitation of
frequency from 0 to 10 Hz and at a dynamic amplitude of 2.5 mm.
Figure 7.34 shows the FFT responses of the body accelerations (bounce and pitch responses) for
the measured and simulated results. The figure shows good agreement in terms of the dominant
peak and corresponding natural frequencies for the suspension equipped with MR damper at wheel
stations 3, 4 and 5. However, the acceleration responses of the suspension system equipped with
224
MR damper at wheel stations 1 and 2 show slight shift in the bounce and pitch natural frequencies
due to the change in the suspension stiffness. Also, the stiffening characteristics of the passive MR
damper, which was discussed previously, is responsible for the increase in the bounce frequency
seen in the measured data particularly for wheel stations 1, 2 and 3. Overall, it can be concluded
that the theoretical model produces quite similar body acceleration responses as the measured
responses.
225
0.16 0.7
Predicted at wh1
0.14 0.6 Measured at wh1
0.12

0.5
Bounce Acc. (m/s2)
0.1
0.4
0.08
0.3
0.06
0.2
0.04
0.02 0.1
0 0
0 2 4 6 8 10 0 2 4 6 8 10
0.09 0.5
Predicted at wh2
0.08 Measured at wh2
0.07 0.4

Bounce Acc. (m/s2)
0.06
0.3
0.05
0.04
0.2
0.03
0.02 0.1
0.01
0 0
0 2 4 6 8 10 0 2 4 6 8 10
0.04 0.2
Predicted at wh3
0.035 Measured at wh3
0.03 0.15
Bounce Acc. (m/s2)
0.025
0.02 0.1
0.015
0.01 0.05
0.005
0 0
0 2 4 6 8 10 0 2 4 6 8 10
0.05 0.35
Predicted at wh4
0.3 Measured at wh4
0.04
0.25
Bounce Acc. (m/s2)
0.03
0.2
0.15
0.02
0.1
0.01
0.05
0 0
0 2 4 6 8 10 0 2 4 6 8 10
0.06 0.25
Predicted Predicted at wh4
Measured Measured at wh5
0.05
0.2
Bounce Acc. (m/s2)
0.04
0.15
0.03
0.1
0.02
0.05
0.01
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Figure ‎7.34 Measured and predicted BA and AA responses of the damped suspension system with
one passive MR damper at different locations under swept sinusoidal excitation
226
7.7.5 Skyhook control validation using one MR damper
Additional validation of the controlled MR test rig with the Simulink model is performed. The
purpose of the validation is to build confidence in using the theoretical model in chapter 8 but with
real vehicle parameters in order to evaluate the performance of the full size tracked vehicle
suspension with MR dampers. In the experiment, the test rig was equipped with one MR damper
located at wheel 4. This wheel station offers enough ‘room’ (gain margin) for the controller to be
effective. The skyhook control with 100 Ns/m gain is used as a system controller while Heaviside
control is used as the MR damper controller. The test rig was excited with a sinusoidal excitation of
3.5 Hz for duration of 10 sec. Then, the bounce and angular accelerations of the body are
measured. In the simulation, the 7DOF model along with the Modified Bouc-Wen model (developed
in chapter 5) of the MR damper are built in Simulink as well as the system and damper controllers.
The test rig parameters such as body and wheel masses and all suspension configurations listed in
Table 6.2 are used in the simulation.
Figure 7.35 shows the time and FFT values of the body accelerations for the measured and
simulated results. The predicted response is for the case of no fictitious viscous damper (C = 0
Ns/m) between wheel mass 3 and the body mass. The figure shows good agreement for the
dominant peak at frequency of 3.5 Hz. There is another peak at about a frequency of 4.3 Hz which
corresponds to the bounce natural frequency of the suspension model. This other peak appears in
the predicted response but not in the measured response. The reason for this discrepancy is due to
the friction effects present in the test rig which is not accounted for directly in the theoretical model.
But when a fictitious viscous damper of damping coefficient C = 80 Ns/m is introduced between
wheel mass 3 and the body mass, to represent the frictional damping present in the experimental
rig, Figure 7.36 shows that there is no more response at the resonance frequency of 4.3 Hz. In
addition, Figure 7.37 shows the comparison between the measured and predicted command
voltage applied to the MR dampers using the skyhook control. The figure shows a good match
between the command voltage produced from both theoretical and experimental controllers.
227
0.6 0.6
Measured
Predicted
0.4 0.5
0.2 0.4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0 0.3
-0.2 0.2
-0.4 0.1
0
0 1 2 3 4 5 0 2 4 6 8 10
15 8
Measured
7 Predicted
10
6

5
5
0 4
3
-5
2
-10
1
-15 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure ‎7.35 Validation of Simulink model and controlled MR suspension rig using skyhook control
(G=100 Ns/m) under sinusoidal excitation of 3.5 Hz and C=0 Ns/m
0.5 0.4
Measured
0.4 Predicted
0.35
0.3
0.3
0.2
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.25
0.1
0 0.2
-0.1
0.15
-0.2
0.1
-0.3
0.05
-0.4
-0.5 0
0 1 2 3 4 5 0 2 4 6 8 10
10 8
Measured
8 Predicted
7
6
6
4
5
2
0 4
-2
3
-4
2
-6
1
-8
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure 7.36 Validation of Simulink model and controlled MR suspension rig using skyhook control
(G=100 Ns/m) under sinusoidal excitation of 3.5 Hz and C=80 Ns/m
228
6 6
Measured Predicted
5 5
4 4
Voltage (v)
Voltage (v)
3 3
2 2
1 1
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
Figure 7.37 Skyhook control command voltage applied to MR damper
7.7.6 Hybrid control validation using one MR damper

Similarly, the controlled results of the MR test rig using hybrid control and the Simulink model are
also compared in this section. In the experiment, one MR damper is mounted at wheel station
number 4. The hybrid control with 50 Ns/m gain and 0.5 weighting factor is used as a system
controller while the Heaviside control is also used as the MR damper controller. The test rig was
excited with the same sinusoidal excitation of 3.5 Hz. Then, the bounce and angular accelerations
of the body are measured. In the simulation, the theoretical 7DOF model of the test rig along with
the Modified Bouc-Wen model of the MR damper are built in Simulink as well as the system and
damper controllers. All the simulation parameters are also taken from Table 6.2.
Figure ‎7.38 shows the comparison between the bounce and angular acceleration responses of the
and test rig under sinusoidal excitation. The predicted response is for the case of no fictitious
viscous damper (C = 0 Ns/m) between wheel mass 3 and the body mass. The figure shows a close
match between the measured and predicted responses in terms of the dominant frequency of 3.5
Hz. However, the model predicts another peak at about a frequency of 4.1 Hz which corresponds
to the bounce natural frequency of the suspension model. This other peak is not present in the
measured response. As noted previously, the reason for this discrepancy is due to the friction
effects present in the test rig which are not accounted for in the theoretical model. When a fictitious
viscous damper of damping coefficient C = 150 Ns/m is introduced between wheel mass 3 and the
body mass, to represent the frictional damping present in the experimental rig, Figure 7.39 shows
that the response at the resonance frequency of 4.1 Hz is suppressed. Furthermore, the command
voltage applied to the MR damper is shown in Figure ‎7.40. It can be seen from the figure that the
measured voltage is the same as the predicted voltage in terms of the maximum value.
It should be noted that the validated Simulink model is used in the simulation of the full size MR
tracked vehicle suspension in chapter 8. In this full size model, the parameters of the real tracked
vehicle along with its suspension configurations are used instead of the scaled test rig parameters.
In addition, the same control methods (skyhook and hybrid) with different gains along with a new
controller, namely fuzzy-hybrid, are used in the semi-active control of the full size tracked vehicle.
The fuzzy-hybrid control is used to simultaneously determine the weighting factor of the hybrid
control which is manually specified in the measured and simulated results of the test rig suspension
presented in this section.
229
0.5 0.9
Measured
0.4 0.8 Predicted
0.3 0.7
0.2
0.6
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.1
0.5
0
0.4
-0.1
0.3
-0.2
0.2
-0.3
-0.4 0.1
-0.5 0
0 1 2 3 4 5 0 2 4 6 8 10
10 9
Measured
8 8 Predicted
6 7
4

6
2
5
0
4
-2
3
-4
2
-6
-8 1
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure ‎7.38 Validation of Simulink model and controlled MR suspension rig using hybrid control
(G=50 Ns/m and =0.5) under sinusoidal excitation of 3.5 Hz and C=0 Ns/m
1.5 0.45
Measured
0.4 Predicted
1
0.35
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0.5 0.3
0.25
0
0.2
-0.5 0.15
0.1
-1
0.05
-1.5 0
0 1 2 3 4 5 0 2 4 6 8 10
Measured
8 8 Predicted
6 7
4
6
2
5
0
4
-2
3
-4
-6 2
-8 1
-10 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure ‎7.39 Validation of Simulink model and controlled MR suspension rig using hybrid control
(G=50 Ns/m and =0.5) under sinusoidal excitation of 3.5 Hz and C=150 Ns/m
230
6 6
Measured Predicted
5 5
4 4
Voltage (v)
Voltage (v)
3 3
2 2
1 1
0 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Time (s) Time (s)
Figure ‎7.40 Hybrid control command voltage applied to MR damper
7.7.7 Validation of skyhook control using two MR dampers

This section presents the validation of the controlled MR test rig using skyhook control of two MR
dampers with the Simulink model under sinusoidal excitation. In the experiment, the test rig was
equipped with two MR dampers located at wheels 1 and 5. The controlled suspension is excited
with sinusoidal excitation of frequency 6 Hz for a duration of 10 sec. The gain of the skyhook
control used during the control process is 2000 Ns/m. Then, the bounce and pitch acceleration
responses of the controlled suspension system are measured. On the other hand, the theoretical
model of the suspension with two MR dampers, the modified Bouc-Wen model for the two MR
dampers, two Heaviside damper controllers as well as two skyhook controllers are built in the
Simulink environment. The parameters of the test rig, which are listed in Table 6.2, are used in the
Simulink model.
Figure 7.41 shows the time and FFT responses of the measured and predicted body accelerations
of the suspension system using skyhook control. The figure shows good match between the
bounce and pitch accelerations of the measured and simulated results in terms of the dominant
peak and the corresponding excitation frequency of 6 Hz. Also, the measured and predicted
command voltage of the MR damper located at the first wheel station is shown in Figure 7.42. The
results show that there is good agreement between the measured and predicted voltage applied to
the MR damper.
231
2 1.4
Measured
1.5 Predicted
1.2
1
1
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.5
0.8
0
0.6
-0.5
0.4
-1
-1.5 0.2
-2 0
0 1 2 3 4 5 0 2 4 6 8 10
25 20
Measured
20 18 Predicted
15 16
10 14

5 12
0 10
-5 8
-10 6
-15 4
-20 2
-25 0
0 1 2 3 4 5 0 2 4 6 8 10
Figure ‎7.41 Validation of computer model and test rig with two MR dampers with skyhook control
(G=2000 Ns/m) under excitation frequency of 6 Hz
6 6
Measured Predicted
5 5
4 4
Voltage (v)
Voltage (v)
3 3
2 2
1 1
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
Figure ‎7.42 Command voltage applied to MR damper located at wheel station 1
7.8 Summary
In this chapter, the dynamic performance of the suspension system with MR dampers has been
investigated experimentally and theoretically. The effect of variation in the suspension design
parameters such as numbers and locations of dampers has been studied and the results are
matched with the simulation data in chapters 3 and 4. Open loop and closed loop semi-active
control methods are applied to the suspension with one MR damper in order to reduce the
suspension vibrations. The effect of the controller gains and weighting factor on the performance of
the suspension system is also discussed. It has been shown that hybrid control offers better
suspension performance than skyhook control. Moreover, real time control of the MR suspension
rig with two MR dampers using both skyhook and hybrid control has been presented. Overall, it has
been shown that the performance of the controlled suspension is better than the performance of
the undamped and MR passive suspension. Finally, the measured performance of the undamped
232
and controlled MR suspensions has been validated against the theoretical Simulink model under
sinusoidal excitation.
233
8 CHAPTER 8: SEMI-ACTIVE CONTROL OF TRACKED VEHICLE SUSPENSION
WITH MR DAMPERS
8.1 Introduction
This chapter provides an evaluation of the semi-active control for a full scale tracked vehicle
suspension with MR dampers using the validated suspension model employed in chapter 7. The
real tracked vehicle parameters listed in Table 3.1 are used instead of the scaled test rig
parameters. Three system control methods in addition to one MR damper controller are used in the
simulation. The first two system controllers are the skyhook and hybrid controls while the third
controller is an intelligent fuzzy-hybrid control. Furthermore, a large scale MR damper model which
suits a full size tracked vehicle application is presented. Finally, the assessment of the passive and
semi-active suspension performance along with the proposed controllers is performed under bump,
sinusoidal and random excitations.
8.2 Semi-active control of tracked vehicle suspension with MR dampers

As stated in the introduction, the validated suspension model used in chapter 7 is employed here
for the evaluation of the full scale tracked vehicle suspension with MR dampers. In the simulation,
the real tracked vehicle parameters listed in Table 3.1 are used. The suspension system of the real
tracked vehicle is the same as the test rig suspension system presented in section 6.7. However,
the real tracked vehicle suspension is equipped with three MR dampers which are located at the
first, second and fifth wheel stations.
In addition, the MR damper model parameters for the large scale MR damper to be used for the full
scale tracked vehicle suspension are different from those used for the scaled tracked vehicle
model. Therefore, the description of the large scale modified Bouc-Wen model and its equations is
presented. This is followed by the development of the fuzzy-hybrid control. It should be noted that
the skyhook and hybrid control methods used in this investigation are the same as presented in
chapter 7 but with different control gains and weighting factor. The controller gain of the skyhook is
[Gs=4000] while the hybrid controller gain and weighting factor are [Gs=10,000, Gg=1000 and
=0.5].
8.2.1 Dynamic model of large scale MR damper

The MR damper model presented in chapter 5 is used to simulate the characteristics of the MR
damper produced by Lord Corporation. This model is used for vibration control of road vehicle
suspension system as in [51, 98]. The damping force produed by this damper model is too low for
the tracked vehicle application. Therefore, this model can not be used in the tracked vehicle
suspension system. It should be noted that the same modified Bouc-Wen model (discussd in
chapter 5) but with different model parameters is employed in [34, 47] to control heavy vehicles and
train suspension systems. The new model parameters which are used in the simulation are listed in
234
Table 8.1. The model simulates the dynamic characteristics of a large scale MR daper that suits
the tracked vehicle application.
Table ‎8.1 Parameters of the MR damper fitted to M113 suspension system [34, 47]
Parameters value parameters value
2100 Ns/m 14000 N/m
3500 Ns/Vm 69500 N/Vm
-2
4690 N/m 3630000 m
-2
28300 Ns/m 3630000 m
295 Ns/Vm A 301
500 N/m n 2
0.143 m 190
8.2.2 Fuzzy-hybrid system control

Fuzzy logic control is a non-traditional control strategy which has attracted the interest of many
researchers. The great capabilities of modelling non-linear systems and the possibilities to deal
with uncertainties inside the models make fuzzy logic more effective in vibration control systems.
Besides, fuzzy logic can be adapted with the conventional control methods. It is not necessarily
that fuzzy logic substitutes the conventional control methods but it can be combined with any
classical controller. The major distinction between the traditional control algorithms and the fuzzy
logic algorithm is that with the traditional methods, it is hard to model the non-linear and complex
systems but it is easier to handle that with fuzzy logic. The fuzzy logic includes three main stages:
the fuzzification, the decision making and the defuzzification. The fuzzification process converts the
crisp data into fuzzy values. The decision making computes the controller output based on the
fuzzy rules. The defuzzification converts back the fuzzy output to crisp values. All the components
comprising the fuzzy logic controller are illustrated in Figure 8.1.
Figure ‎8.1 Schematic drawing of fuzzy logic controller
The developed control scheme used in this investigation is a fuzzy-hybrid controller. This controller
has many advantages as it combines the hybrid control effects with the fuzzy logic. The formation
of hybrid controller is already implemented in section 6.6.2. It is the controller gains Gs , G that are
changed to adapt the proposed model. Moreover, the skyhook controller improves the ride comfort
of a vehicle while the groundhook controller increases the vehicle stability and handling. Hence, the
fuzzy-hybrid may improve the vehicle dynamics and increase its mobility. To establish the
235
suggested control, the hybrid controller rule should be adopted. The controlled force of the hybrid
th
control of the i damper is expressed by the following equation:
(‎8.1)
The weighting factor in Eq. 8.1 plays a crucial role in the hybrid control policy and this was
experimentally proved for the scaled model in section 7.4.3. It is very important that the weighting
factor varies depending on the body and the wheel motion. To optimize the control performance,
the weighting factor is adjusted automatically by using the fuzzy logic algorithm. The fuzzy input
variables used in this controller are the vertical velocities across the damper end, θ and
respectively while the fuzzy output variable is the weighting factor .
The membership functions of the two inputs and single output variables of the fuzzy logic controller
are taken as in [72] but with a little change to suit this application. These membership functions are
expressed as follows:
The abbreviations used in the membership functions correspond to:
S= very small, SM =small, M =medium, LA =large and L = very large, = zero
, =0.3, =0.5, =0.7, =0.9, =1
The ranges for the three fuzzy variables are determined based on the passive suspension
dynamics and they take values of [0, 4], [0, 6] and [0, 1] respectively. The possible membership
function graphs of the three fuzzy variables are illustrated below:
VS SM ME L VL
0.8
Degree of membership
0.6
0.4
0.2
0
0 1 2 3 4
BV
Figure 8.2 First input membership function ‘Body velocity’
236
VS SM ME L VL
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6
WV
Figure 8.3 Second input membership function ‘Wheel velocity’

ZE tau1 tau2 tau3 tau4 tau5 tau6
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
tau
Figure 8.4 Output membership function ‘Weighting factor’
0.8
0.6
tau
0.4
0.2 4
3
0
6 2
5
4
3 1
2
1
0 0
BV
WV
Figure 8.5 Surface viewer of the Fuzzy logic control
The fuzzy controller rules are listed in Table 8.2. There are 25 rules adopted for the proposed
controller. ach rule is executed based on the ‘IF-TH N’ statement. For instance, rule number
denoted by (R5) and rule number 21 (R21) are executed as follows:
R5: IF ( = VL) and ( wi =VS) THEN ( =1)
R21: IF ( = VS) and ( wi =VL) THEN ( =0)
237
Table ‎8.2 Fuzzy control rules
Wheel Body velocity
velocity VS SM ME LA VL
VS 3 5 5 6 6
SM 1 3 4 5 5
ME 1 2 3 4 5
LA 1 1 2 3 4
VL ZE 1 2 2 3
The control scheme of the fuzzy-hybrid control along with the suspension model is illustrated in
Figure 8.6. The fuzzy logic controller takes the body velocity (BV) and the wheel velocity (WV)
across each damper end and specifies the appropriate weighting factor ( i ) which is sent to the
hybrid controller. Based on the value of the weighting factor, the hybrid controller specifies the
desired damping force ( ). In order to perform the simulation of the suspension system using the
fuzzy-hybrid control, a Simulink model of the controller should be created. The Simulink model is a
combination of both the hybrid model and fuzzy logic model as shown in Figure 8.7. The values for
the skyhook and groundhook gains used in the simulation are 40,000 and 2000 Ns/m.
Figure ‎8.6 Block diagram of 7-DOF suspension model integrated with fuzzy-hybrid controller
Figure 8.7 Fuzzy-hybrid controller Simulink model

238
8.3 Road excitation profiles
This section presents the road excitations used in the simulation of semi-active suspension with
MR dampers. Three road profiles which are bump, sinusoidal and random excitations are utilized
for the evaluation of the dynamic performance of suspension system. The bump road profile used
in this study is as in reference [101] with bump height of 0.4 m and width of 1.67 m. This type is
used to quantify the transient response characteristics the reader can refer to section 3.3 for more
details. The second type is the sinusoidal road profile which is more realistic to most of the tracked
vehicles. This road profile is used in reference [24] and discussed in section 4.2. The sinusoidal
excitation used in the simulation has wavelength of 5 m and height of 0.1m. Finally, the random
road profile is usually used to evaluate the frequency response. The profile is expressed by Eq. 8.2
which relates the amplitude of excitation and its derivative to the road irregularities and vehicle
speed. The white noise is σ ρ and ρ is the road roughness parameter and it depends upon
-
the road condition [77]. The values of these parameters are the road roughness ( ) and
covariance of the road (σ 3 mm2 ). Figure 8.8 shows the graph of the three road profiles used in
this simulation at vehicle speed 10 km/h.
ρ (‎8.2)
0.4 0.1
Road amplitude (m)
Road amplitude (m)
0.3 0.05
0.2 0
0.1 -0.05
0 -0.1
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
0.2
Road amplitude (m)
0.1
-0.1
-0.2
0 1 2 3 4 5
Time (s)
Figure ‎8.8 Typical bump, sinusoidal and random road profiles imposed at first wheel station
239
8.4 Semi-active control results
In this section, the results from the simulation are presented for the bump, sinusoidal and random
excitations. The vehicle crosses the road profile with three different speeds of 10, 20 and 40 km/h.
These speeds are chosen to be close to the real vehicle speeds over rough terrains. The semi-
active suspension performance is assessed based on two main criteria which are ride comfort and
vehicle handling [72, 102]. The first is closely related to the body accelerations (bounce and pitch)
while the second is certified by wheel bounce. The goal is to improve these two performance
indices by reducing the bounce and angular accelerations of body as well as the wheel bounce.
The suspension performance is studied for five cases of vibration control.
a- Conventional passive i.e. constant damping coefficient of 22520 Ns/m (Table 3.1)
b- Passive MR damper i.e. no current or voltage is applied to the damper.
c- Semi-active control with skyhook control using Equation (6.4).
d- Semi-active control with hybrid control using Equation (6.5)
e- Semi-active control with fuzzy-hybrid control using Equation (8.1)
The above performance criteria are used to quantify the relative performances of these control
methods and to compare with the baseline conventional passive suspension performance.
8.4.1 System responses under bump excitation

In this section, the evaluation of the suspension performance with the proposed controllers under
bump road profiles is discussed. The vehicle crosses the road profile with three constant speeds of
10, 20 and 40 km/h. The assessment of the suspension system performance is studied based on
the body and wheel acceleration responses which are related to the ride comfort and handling
characteristics.
The time histories of the bounce acceleration (BA) and angular acceleration (AA) of the suspension
system under bump road excitation at different speeds are shown in Figure 8.9(a, b). The figure
shows the comparison between the controlled semi-active suspension with skyhook , hybrid and
fuzzy-hybrid controllers in addition to MR passive and the conventional passive system. It is clear
that the BA and AA amplitudes are substantially reduced for all suspension system compared with
the passive suspension. In addition, the BA and AA responses of the passive suspension are
significantly increased when the vehicle crosses the bump road profile with 20 and 40 km/h.
However, for all speeds the controlled suspensions have better isolation than the conventional
passive system. On the other hand, the conventional passive system has a shorter settling time
than the MR off, skyhook and hybrid control systems.
The results can be also seen from Table ‎8.3 which lists the peak-to-peak (PTP) values of the BA
and AA for all suspension systems. The fuzzy-hybrid control is effective for controlling the BA of the
body; it give about 33%, 69% and 56% improvement in terms of vibration reduction for semi-active
control compared to passive. This is followed by hybrid, skyhook and MR passive suspension
systems. In addition, the MR passive and the semi-active controllers are superior to the
conventional passive in reducing the angular acceleration amplitudes.
240
In terms of the road handling characteristics, the wheel bounce responses are investigated for the
system responses under the same excitations. Figure ‎8.10 shows the wheel bounce displacement
of the five road wheels of the suspension systems at a speed of 10 km/h. Overall, the semi-active
control suspension has reduced the settling time and dissipated the energy of the wheel
displacement compared with the conventional passive system. The performance of the controlled
suspension can be evaluated from Table ‎8.4. The semi-active controls with skyhook , hybrid and
fuzzy-hybrid have reduced the bounce displacement of the first road wheel by about 5%. Also, the
bounce displacement of the last road wheel is reduced by about 2.5%, 3.1% and 5.0%,
respectively, by these three semi-active control techniques. However, the bounce displacement of
the intermediate road wheels (second, third and fourth), which have no dampers, is slightly
increased compared to the conventional passive system except for the case of the fuzzy-hybrid
control that reduces the response of the fourth wheel by 1.7%.
Figure 8.11 shows the wheel bounce displacement of the suspension system at 20 km/h. The
fuzzy-hybrid control is superior in improving the bounce displacement of the road wheels with MR
dampers than the rest of the semi-active control systems. However, the MR off, skyhook and hybrid
control methods have reduced the bounce displacement of the first and last wheels only as can be
seen from Table 8.5.
Figure ‎8.12 shows the time histories of wheel responses when the vehicle crosses the bump terrain
with a speed of 40 km/h. The suspension performance with the conventional passive dampers is
better than the semi-active control suspension. It can be noticed from Table ‎8.6 that the PTP
values of wheel bounce of the semi-active suspicion is increased while the passive suspension
offer better road handling at a specific speed.
The damping forces and the corresponding command voltage for the MR damper mounted on the
first wheel station are shown in Figure 8.13. It can be seen that the damping force of the MR
damper can follow the controlled (desired) force. Also, the damping force of MR off (0 volt) is lower
than the damping force of the fuzzy-hybrid control. Moreover, the fuzzy-hybrid is more reliable in
terms of the power consumption (i.e. less time of the applied voltage) used compared to the
skyhook and hybrid controllers.
241
Speed 10 km/h
10 8
Passive Passive
8 MR off MR off
Skyhook 6 Skyhook
6 Hybrid Hybrid
Fuzzy-Hybrid 4 Fuzzy-Hybrid
4

Bounce Acc (m/s2)
2
2
0 0
-2
-2
-4
-4
-6
-6
-8
-10 -8
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
Speed 20 km/h
15 15
Passive Passive
MR off MR off
10
Skyhook Skyhook
Hybrid 10 Hybrid
5 Fuzzy-Hybrid Fuzzy-Hybrid
0
Bounce Acc (m/s2)
5
-5
-10
0
-15
-20
-5
-25
-30 -10
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
Speed 40 km/h
30 15
Passive Passive
MR off MR off
Skyhook Skyhook
20 10
Hybrid Hybrid
Fuzzy-Hybrid Fuzzy-Hybrid
10 5
Bounce Acc (m/s2)
0 0
-10 -5
-20 -10
-30 -15
0 1 2 3 4 5 0 1 2 3 4 5
Time (s) Time (s)
(a) (b)
Figure ‎8.9 Time histories of system responses under bump excitation at different speeds
242
0.6
Passive
0.5 MR off
Skyhook
Wheel# 1 Disp (m)

0.4 Hybrid
Fuzzy-Hybrid
0.3
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 2 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 3 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 4 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.6
Passive
0.5 MR off
Skyhook
Wheel# 5 Disp (m)
0.4 Hybrid
Fuzzy-Hybrid
0.3
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
Figure ‎8.10 Time histories of wheels bounce displacement under bump excitation; speed 10 km/h
243
0.4
Passive
MR off
0.3 Skyhook
Wheel# 1 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 2 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 3 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.4
Passive
MR off
0.3 Skyhook
Wheel# 4 Disp (m)
Hybrid
0.2 Fuzzy-Hybrid
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.6
Passive
0.5 MR off
Skyhook
Wheel# 5 Disp (m)
0.4 Hybrid
Fuzzy-Hybrid
0.3
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
Figure ‎8.11 Time histories of wheels displacement under bump excitation; speed 20 km/h
244
0.5
Passive
0.4 MR off
Skyhook
Wheel# 1 Disp (m)

Hybrid
0.3
Fuzzy-Hybrid
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.5
Passive
0.4 MR off
Skyhook
Wheel# 2 Disp (m)
Hybrid
0.3
Fuzzy-Hybrid
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.5
Passive
0.4 MR off
Skyhook
Wheel# 3 Disp (m)
Hybrid
0.3
Fuzzy-Hybrid
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.5
Passive
0.4 MR off
Skyhook
Wheel# 4 Disp (m)
Hybrid
0.3
Fuzzy-Hybrid
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
0.5
Passive
0.4 MR off
Skyhook
Wheel# 5 Disp (m)
Hybrid
0.3
Fuzzy-Hybrid
0.2
0.1
-0.1
0 1 2 3 4 5
Time (s)
Figure ‎8.12 Time histories of wheels displacement under bump excitation; speed 40 km/h
245
4
x 10
5
MR off
Desired Force
4 Fuzzy-Hybrid control
2
Damping force (N)
-1
-2
-3
0 1 2 3 4 5
Time (s)
14
Skyhook
Hybrid
Fuzzy-Hybrid
12
10
8
Voltage (V)
0
2.011 2.0113 2.0115
Time (s)
Figure ‎8.13 Damping forces and command voltage of various control methods for MR damper
mounted on the first wheel under bump excitation; speed 10km/h
246
Table ‎8.3 PTP values and improvement of body acceleration measures under bump road excitation and various speeds
Improvement w.r.t conventional. passive (%)
Suspension Speed Conv. skyhook hybrid fuzzy-hybrid
MR Off skyhook hybrid fuzzy-hybrid
index (km/h) Passive Control Control Control MR Off
Control Control Control
2
10 17.646 13.933 12.590 12.182 11.895 21.0 28.7 31.0 32.6
BA (m/s ) 20 40.242 14.535 14.531 14.184 12.590 63.9 63.9 64.8 68.7
40 56.800 31.922 29.175 28.093 25.003 43.8 48.6 50.5 56.0
2
10 11.918 14.614 12.411 11.767 8.969 -22.6 -4.1 1.3 24.7
AA (rad/s ) 20 22.249 13.982 12.732 12.337 10.259 37.2 42.8 44.6 53.9
40 26.726 13.799 13.754 13.799 14.311 48.4 48.5 48.4 46.5
Table ‎8.4 PTP values and improvement of wheel displacement measures under bump road excitation and 10 km/h
Improvement w.r.t conventional passive (%)
Suspension Conv. skyhook hybrid fuzzy-hybrid
index Passive Control Control Control MR Off
Wheel 1 0.443 0.415 0.421 0.422 0.421 6.3 5.0 4.9 5.1
Wheel 2 0.380 0.412 0.406 0.405 0.400 -8.3 -6.7 -6.4 -5.1
Wheel 3 0.369 0.370 0.371 0.372 0.373 -0.3 -0.5 -0.7 -1.0
Wheel 4 0.386 0.407 0.397 0.395 0.380 -5.4 -2.9 -2.3 1.7
Wheel 5 0.468 0.468 0.456 0.454 0.445 0.0 2.5 3.1 5.0
247
Wheel 1 0.445 0.415 0.410 0.409 0.400 6.7 7.9 8.1 10.2
Wheel 2 0.407 0.412 0.412 0.410 0.399 -1.2 -1.2 -0.7 2.0
Wheel 3 0.401 0.412 0.412 0.411 0.404 -2.7 -2.7 -2.5 -0.7
Wheel 4 0.387 0.410 0.403 0.401 0.393 -5.9 -4.1 -3.6 -1.6
Wheel 5 0.473 0.405 0.395 0.393 0.397 14.4 16.5 16.9 16.1
Wheel 1 0.373 0.475 0.474 0.473 0.460 -27.6 -27.1 -26.8 -23.4
Wheel 2 0.377 0.477 0.479 0.479 0.484 -26.7 -27.2 -27.2 -28.4
Wheel 3 0.535 0.546 0.542 0.540 0.526 -2.0 -1.3 -0.9 1.8
Wheel 4 0.523 0.548 0.539 0.536 0.526 -4.9 -3.0 -2.6 -0.7
Wheel 5 0.346 0.479 0.484 0.485 0.490 -38.4 -39.7 -40.1 -41.3
248
8.4.2 System responses under sinusoidal excitation
The effectiveness of the semi-active suspension with MR dampers is quantified when the vehicle
crosses a sinusoidal road excitation with different speeds. Figure 8.14(a, b) shows the time and
frequency domains of the bounce acceleration of the body under the given excitations. A
reasonable improvement in the BA responses is achieved at low speed for the suspension with MR
off and semi-active control compared to the conventional passive suspension. The suspension
performance with skyhook and hybrid control is better than the suspension with Fuzz-Hybrid control
at this specific speed. However, at 20 and 40 km/h speed, the suspension with the MR off and
semi-active control methods are not as desired. These results can be also verified from Table 8.7.
The FFT responses show that at low speed, the controlled suspension system isolates the body
from the road vibration in the frequency range from 2 Hz to 12 Hz. The reason for the deterioration
in the BA responses is the insufficient dynamic force generated by the MR damper model. It is
noted that the dynamic damping force of both the hydraulic and the MR damper models are
assumed to be the same but it seems that the MR damper model has low damping force.
Similarly, Figure 8.15(a, b) shows the AA responses of the suspension under the same excitations
and vehicle speeds. At 10 and 20 km/h, the suspension with fuzzy-hybrid control reduces the AA
responses while the suspension with MR off, skyhook and hybrid control degrade the AA response
compared to the conventional passive suspension. On the other hand, the AA responses of the MR
off and controlled suspension are greatly reduced compared to the conventional passive case
when the vehicle crosses the road profile with a speed of 40 km/h. The effectiveness of the
controlled suspension can be quantified from the FFT responses of the AA which are shown in
Figure 8.15(b). The figure shows that the controlled suspension reduces the peak amplitudes of the
AA around the pitch and bounce resonant frequencies.
Like the bump excitation analysis, the road handling characteristics which are related to the wheel
bounce responses are investigated for the system responses under the sinusoidal excitations.
Figure ‎8.16 shows the wheel bounce displacement of the five road wheels of the suspension
systems under the periodic excitation with a vehicle speed of 10 km/h. Overall, the bounce
displacement of the first wheel of the controlled suspension is reduced. However, the controlled
suspensions increase the other wheel bounce except for the fuzzy-hybrid control which keeps the
wheels bounce almost unchanged. The performance of the controlled suspension can be evaluated
from the RMS values of the FFT responses listed in Table ‎8.8. The fuzzy-hybrid control has
reduced the bounce displacement of the first and fourth road wheels by about 1.7% and 0.5%.
Figure 8.17 shows the wheel bounce displacements of the suspension system under the same
excitation and vehicle speed of 20 km/h. The results show much better improvement in the wheel
bounce displacement for the MR off and semi-actively controlled suspensions. However, the
intermediate wheel responses of the conventional passive suspension still have better performance
than the semi-actively controlled suspensions. The peak amplitudes of the FFT responses of the
semi-actively controlled and the MR off suspensions are reduced in the neighbourhood of the body
resonant frequencies (0.9 Hz and 1.5 Hz) while the peak amplitudes at the wheel resonant
frequency (12.6 Hz) remain unchanged. The suspension performance can be evaluated from
249
Table 8.9 where the MR off and all the semi-active controllers have reduced the wheel responses
except the third wheel response which is increased.
Figure ‎8.18 shows the time histories of the wheel responses when the vehicle crosses the
sinusoidal terrain with a speed of 40 km/h. The suspension with MR off and semi-active control
have better dynamic behaviour than the conventional passive suspension. The bounce
displacements of the wheels have been reduced. It can be noticed from Table ‎8.10 that the RMS
values of wheels bounce of the semi-active suspension are reduced around the excitation
frequency of 2 Hz.
The damping forces and the corresponding command voltage for MR damper mounted on the first
wheel station are shown in Figure 8.19. It can be seen that the damping force of the MR damper
can follow the controlled (desired) force. Also, the damping force of MR off (0 volt) is lower than the
damping force of the fuzzy-hybrid control.
250
Speed 10 km/h
8 1.4
Passive Passive
MR off MR off
Skyhook Skyhook
1.2
6 Hybrid Hybrid
1
4
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0.8
0.6
0
0.4
-2
0.2
-4 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 20 km/h
8 4
Passive Passive
MR off MR off
6 Skyhook 3.5 Skyhook
Hybrid Hybrid
4
3
2
2.5
Bounce Acc (m/s2)
Bounce Acc (m/s2)

0
2
-2
1.5
-4
1
-6
-8 0.5
-10 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 40 km/h
15 6
Passive Passive
MR off MR off
Skyhook Skyhook
10 Hybrid 5 Hybrid
5 4
Bounce Acc (m/s2)
Bounce Acc (m/s2)
0 3
-5 2
-10 1
-15 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎8.14 Bounce acceleration responses of suspension systems with different control methods
under sinusoidal excitation at different speeds
251
Speed 10 km/h
14 1.8
Passive Passive
MR off MR off
12 1.6
Skyhook Skyhook
Hybrid Hybrid
1.4
8
1.2

6
1
4
0.8
2
0.6
0
0.4
-2
-4 0.2
-6 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 20 km/h
15 7
Passive Passive
MR off MR off
Skyhook Skyhook
6
10 Hybrid Hybrid
5
5
-5
2
-10
1
-15 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
Speed 40 km/h
15 9
Passive Passive
MR off MR off
Skyhook 8 Skyhook
Hybrid Hybrid
7
6
5
5
4
0
2
-5
-10 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎8.15 Angular acceleration responses of suspension systems with different control methods
under sinusoidal excitation at different speeds
252
0.15 0.12
Passive Passive
0.1 MR off MR off
0.1
Skyhook Skyhook
Hybrid Hybrid
0.05
Wheel# 1 Disp (m)
Wheel# 1 Disp (m)

Fuzzy-Hybrid 0.08 Fuzzy-Hybrid
0
0.06
-0.05
0.04
-0.1
-0.15 0.02
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.1
Passive Passive
0.1 MR off MR off
Skyhook 0.08 Skyhook
Hybrid Hybrid
0.05
Wheel# 2 Disp (m)
Wheel# 2 Disp (m)

0.06
0
-0.05
0.04
-0.1
0.02
-0.15
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.09
Passive Passive
MR off 0.08 MR off
0.1
Skyhook Skyhook
0.07
Hybrid Hybrid
0.05
Wheel# 3 Disp (m)
Fuzzy-Hybrid Wheel# 3 Disp (m) Fuzzy-Hybrid

0.06
0 0.05
-0.05 0.04
0.03
-0.1
0.02
-0.15
0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.2 0.1
Passive Passive
0.15 MR off MR off
0.1 Hybrid Hybrid
Wheel# 4 Disp (m)
Wheel# 4 Disp (m)
0.05 0.06
0
-0.05 0.04
-0.1
0.02
-0.15
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.12
Passive Passive
0.1 MR off MR off
0.1
Skyhook Skyhook
Hybrid Hybrid
0.05
Wheel# 5 Disp (m)
Wheel# 5 Disp (m)
0
0.06
-0.05
0.04
-0.1
-0.15 0.02
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
Figure ‎8.16 Wheel displacement responses of suspension systems with different control methods
under sinusoidal excitation at speed 10 km/h
253
0.15 0.12
Passive Passive
0.1 MR off MR off
0.1
Skyhook Skyhook
Hybrid Hybrid
0.05
Wheel# 1 Disp (m)
Wheel# 1 Disp (m)

0
0.06
-0.05
0.04
-0.1
-0.15 0.02
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.1
Passive Passive
0.1 MR off MR off
Hybrid Hybrid
0.05
Wheel# 2 Disp (m)
Wheel# 2 Disp (m)

0.06
0
-0.05
0.04
-0.1
0.02
-0.15
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.09
Passive Passive
MR off 0.08 MR off
0.1
Skyhook Skyhook
0.07
Hybrid Hybrid
0.05
Wheel# 3 Disp (m)
Wheel# 3 Disp (m)
0 0.05
-0.05 0.04
0.03
-0.1
0.02
-0.15
0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.2 0.08
Passive Passive
0.15 MR off 0.07 MR off
Skyhook Skyhook
0.1 Hybrid 0.06 Hybrid
Wheel# 4 Disp (m)
Wheel# 4 Disp (m)
0.05 0.05
0 0.04
-0.05 0.03
-0.1 0.02
-0.15 0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.12
Passive Passive
0.1 MR off MR off
0.1
Skyhook Skyhook
Hybrid Hybrid
0.05
Wheel# 5 Disp (m)
Wheel# 5 Disp (m)
0
0.06
-0.05
0.04
-0.1
-0.15 0.02
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
254
0.15 0.1
Passive Passive
0.1 MR off MR off
Hybrid Hybrid
0.05
Wheel# 1 Disp (m)
Wheel# 1 Disp (m)

0.06
0
-0.05
0.04
-0.1
0.02
-0.15
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.1
Passive Passive
0.1 MR off MR off
Hybrid Hybrid
0.05
Wheel# 2 Disp (m)
Wheel# 2 Disp (m)

0.06
0
-0.05
0.04
-0.1
0.02
-0.15
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.09
Passive Passive
MR off 0.08 MR off
0.1
Skyhook Skyhook
0.07
Hybrid Hybrid
0.05
Wheel# 3 Disp (m)
Wheel# 3 Disp (m)

0 0.05
-0.05 0.04
0.03
-0.1
0.02
-0.15
0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.2 0.08
Passive Passive
0.15 MR off 0.07 MR off
Skyhook Skyhook
0.1 Hybrid 0.06 Hybrid
Wheel# 4 Disp (m)
Wheel# 4 Disp (m)
0.05 0.05
0 0.04
-0.05 0.03
-0.1 0.02
-0.15 0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
0.15 0.09
Passive Passive
MR off 0.08 MR off
0.1
Skyhook Skyhook
0.07
Hybrid Hybrid
0.05
Wheel# 5 Disp (m)
Wheel# 5 Disp (m)
0 0.05
-0.05 0.04
0.03
-0.1
0.02
-0.15
0.01
-0.2 0
0 1 2 3 4 5 0 2 4 6 8 10 12 14
(a) (b)
255
4
x 10
2
MR off
Desired Force
1.5 Fuzzy-Hybrid control
0.5
Damping force (N)
-0.5
-1
-1.5
-2
-2.5
0 1 2 3 4 5
Time (s)
14
Skyhook
Hybrid
Fuzzy-Hybrid
12
10
8
Voltage (V)
0
1.26 1.261 1.262
Time (s)
mounted on the first wheel under sinusoidal excitation; speed 10km/h
256
Table ‎8.7 RMS values and improvement of BA and AA measures under sinusoidal road excitation
Conv. skyhook hybrid fuzzy-hybrid Improvement w.r.t conventional passive (%)

Suspension Speed MR Off
Passive -3 Control Control Control skyhook hybrid fuzzy-hybrid
index (km/h) -3 (x10 ) -3 -3 -3 MR Off
(x10 ) (x10 ) (x10 ) (x10 ) Control Control Control
2
10 2.31 2.30 2.24 2.22 2.27 0.4 3.0 3.9 1.7
BA (m/s ) 20 3.17 7.30 6.35 6 5.03 -130.3 -100.3 -89.3 -58.7
40 5.99 10.61 10.58 10.56 7.56 -77.1 -76.6 -76.3 -26.2
2
10 3 4.34 3.44 3.3 2.55 -44.7 -14.7 -10.0 15.0
AA (rad/s ) 20 9.68 12.59 10.79 10.37 7.11 -30.1 -11.5 -7.1 26.5
40 14.76 5.51 5.43 5.43 5.8 62.7 63.2 63.2 60.7
Table ‎8.8 RMS values and improvement of wheel displacement measures under sinusoidal road excitation and 10 km/h

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Off
Wheel 1 0.178 0.177 0.176 0.176 0.175 0.6 1.1 1.1 1.7
Wheel 2 0.156 0.160 0.106 0.160 0.158 -2.6 -2.6 -2.6 -1.3
Wheel 3 0.147 0.149 0.149 0.149 0.147 -1.4 -1.4 -1.4 0.0
Wheel 4 0.183 0.186 0.185 0.184 0.182 -1.6 -1.1 -0.5 0.5
Wheel 5 0.170 0.177 0.175 0.174 0.170 -4.1 -2.9 -2.4 0.0
257

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Passive
Wheel 1 0.196 0.099 0.109 0.114 0.162 49.5 44.4 41.8 17.3
Wheel 2 0.170 0.147 0.151 0.153 0.161 13.5 11.2 10.0 5.3
Wheel 3 0.142 0.157 0.155 0.153 0.143 -10.6 -9.2 -7.7 -0.7
Wheel 4 0.169 0.163 0.164 0.164 0.169 3.6 3.0 3.0 0.0
Wheel 5 0.198 0.097 0.110 0.115 0.168 51.0 44.4 41.9 15.2

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Off
Wheel 1 0.166 0.130 0.130 0.130 0.130 21.7 21.7 21.7 21.7
Wheel 2 0.162 0.131 0.131 0.130 0.138 19.1 19.1 19.8 14.8
Wheel 3 0.147 0.136 0.136 0.136 0.139 7.5 7.5 7.5 5.4
Wheel 4 0.170 0.167 0.167 0.167 0.169 1.8 1.8 1.8 0.6
Wheel 5 0.155 0.139 0.141 0.141 0.148 10.3 9.0 9.0 4.5
258
8.4.3 System responses under random excitation
The effectiveness of the semi-active suspension with MR dampers is finally quantified when the
vehicle crosses a random road excitation with different speeds. Figure 8.20 shows the power
spectral density (PSD) of the body bounce acceleration under the given excitations for passive and
semi-active control. At 10 km/h, the conventional passive suspension has a better isolation than
MR off, skyhook and hybrid controllers in the vicinity of the first resonant frequency. However, the
fuzzy-hybrid controller is better than the other controllers in attenuating the first resonance peak.
On the other hand, for attenuating the frequency responses other than the first resonance, all the
proposed semi-active controllers are superior to the conventional passive suspension.
At speed 20 km/h, the semi-active controllers have significantly reduced the frequency responses
compared to the conventional passive one. On the other hand, at relatively high speed of 40 km/h
(this is a practical maximum speed that can be reached on that particular random road) the
conventional passive has better response than the semi-active controllers at the first peak
frequency response. However, the fuzzy-hybrid controller is superior to all the other controllers in
reducing the frequency responses other than at the first resonance frequency. These results can
also be assessed from Table 8.11 which shows the RMS values of the BA of all suspension
systems at various speeds.
Similarly, the PSD of the AA responses under the described excitation are shown in Figure 8.21. It
can be seen that the AA responses of the semi-active suspension are reduced compared to the
conventional passive suspension. However, the passive suspension has a better isolation at the
first peak frequency responses. The results shown in Table 8.11 confirm that the fuzzy-hybrid is still
superior to all other controllers in improving the AA responses at all speeds. The fuzzy-hybrid
controller has reduced the RMS values of the AA at speeds 10, 20 and 40 km/h by about 15%,
32% and 67% respectively.
Like the sinusoidal excitation analysis, the road handling characteristics which are related to the
wheel bounce displacements are investigated for the system responses under the random
excitations. Figure ‎8.22 to Figure ‎8.24 show the PSD of all wheel displacements at various vehicle
speeds. It can be seen that the semi-active suspensions have relatively better isolation at the first
peak frequency than the conventional passive suspension. However, the conventional passive
suspension has improved the RMS values of the wheels displacement at the wheel resonant
frequency compared to the semi-actively controlled suspension. These results can be drawn from
Table ‎8.12 to Table ‎8.14. It can be noted that the semi-actively controlled suspension system did
not offer much more improvement in the wheel displacements.
The damping forces and the corresponding command voltage for the MR damper mounted on the
first wheel station are shown in Figure 8.25. It can be seen that the damping force of the MR
damper can follow the controlled (desired) force specified by the fuzzy-hybrid controller. Also, the
damping force of MR off (0 volt) is lower than the damping force of the fuzzy-hybrid control.
259
Speed 10 km/h
1
10
Passive
MR off
Skyhook
Hybrid
0 Fuzzy-Hybrid
10
PSD Bounce Acc ((m/s2)2/ Hz)
-1
10
-2
10
-3
10 0 1
10 10
Frequency (Hz)
Speed 20 km/h
2
10
Passive
MR off
Skyhook
Hybrid
Fuzzy-Hybrid
1
10
0
10
-1
10
-2
10 0 1
10 10
Frequency (Hz)
Speed 40 km/h
3
10
Passive
MR off
Skyhook
2
Hybrid
10 Fuzzy-Hybrid
1
10
0
10
-1
10
-2
10 0 1
10 10
Frequency (Hz)
Figure ‎8.20 Bounce acceleration responses of suspension systems with different control methods
under random excitation at different speeds
260
Speed 10 km/h
0
10
Passive
MR off
Skyhook
Hybrid
Fuzzy-Hybrid
-1
10
PSD Angular Acc ((rad/s2)2/ Hz)
-2
10
-3
10
-4
10 0 1
10 10
Frequency (Hz)
Speed 20 km/h
1
10
Passive
MR off
Skyhook
Hybrid
Fuzzy-Hybrid
0
10
-1
10
-2
10
-3
10 0 1
10 10
Frequency (Hz)
Speed 40 km/h
2
10
Passive
MR off
Skyhook
Hybrid
Fuzzy-Hybrid
1
10
0
10
-1
10
-2
10 0 1
10 10
Frequency (Hz)
Figure ‎8.21 Angular acceleration responses of suspension systems with different control methods
under random excitation at different speeds
261
-2
10
Passive
MR off
-3
10 Skyhook
Hybrid
Fuzzy-Hybrid
PSD Wheel#1 Disp (m 2/Hz)
-4
10
-5
10
-6
10
-7
10
-8
10
0 1
10 10
Frequency (Hz)
-2
10
Passive
MR off
-3
10 Skyhook
Hybrid
Fuzzy-Hybrid
-4
10
-5
10
-6
10
-7
10
-8
10
0 1
10 10
Frequency (Hz)
-2
10
Passive
MR off
Skyhook
Hybrid
-3
10 Fuzzy-Hybrid
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
-2
10
Passive
MR off
Skyhook
Hybrid
-3
10 Fuzzy-Hybrid
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
-2
10
Passive
MR off
-3
10 Skyhook
Hybrid
Fuzzy-Hybrid
-4
10
-5
10
-6
10
-7
10
-8
10
0 1
10 10
Frequency (Hz)
under random excitation at speed 10 km/h
262
-1
10
Passive
MR off
-2
10 Skyhook
Hybrid
Fuzzy-Hybrid

-3
10
-4
10
-5
10
-6
10
-7
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
-2
10 Skyhook
Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
-7
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
Skyhook
Hybrid
-2
10 Fuzzy-Hybrid
-3
10
-4
10
-5
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
Skyhook
-2
10 Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
-2
10 Skyhook
Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
-7
10
0 1
10 10
Frequency (Hz)
263
-1
10
Passive
MR off
Skyhook
-2
10 Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
Skyhook
-2
10 Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
0
10
Passive
MR off
Skyhook
-1
10 Hybrid
Fuzzy-Hybrid
-2
10
-3
10
-4
10
-5
10
0 1
10 10
Frequency (Hz)
0
10
Passive
MR off
Skyhook
-1
10 Hybrid
Fuzzy-Hybrid
-2
10
-3
10
-4
10
-5
10
0 1
10 10
Frequency (Hz)
-1
10
Passive
MR off
Skyhook
-2
10 Hybrid
Fuzzy-Hybrid
-3
10
-4
10
-5
10
-6
10
0 1
10 10
Frequency (Hz)
264
4
x 10
1.5
MR off
Desired Force
Fuzzy-Hybrid control
Damping force (N)

0.5
-0.5
-1
0 1 2 3 4 5
Time (s)
14
Skyhook
Hybrid
Fuzzy-Hybrid
12
10
8
Voltage (V)
0
3.0612 3.0616 3.062
Time (s)
mounted on the first wheel under random excitation; speed 10km/h
265
Table ‎8.11 RMS values and improvement of BA and AA measures under random road excitation

Suspension Speed Conv. MR skyhook hybrid fuzzy-hybrid
index (km/h) Passive Off Control Control Control skyhook hybrid fuzzy-hybrid
MR Off
2
10 0.304 0.309 0.305 0.304 0.290 -1.4 -0.3 0.1 4.8
BA (m/s ) 20 0.64 0.118 0.117 0.116 0.125 81.6 81.7 81.9 80.5
40 13.188 24.367 18.139 15.711 12.154 -84.8 -37.5 -19.1 7.8
2
10 0.013 0.028 0.024 0.023 0.011 -115.4 -84.6 -76.9 15.4
AA (rad/s ) 20 0.136 0.327 0.265 0.236 0.092 -140.4 -94.9 -73.5 32.4
40 1.712 0.913 0.783 0.72 0.571 46.7 54.3 57.9 66.6
Table ‎8.12 RMS values and improvement of wheel displacement measures under random road excitation and 10 km/h

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Off
Wheel 1 0.101 0.101 0.1 0.1 0.099 0.0 1.0 1.0 2.0
Wheel 2 0.088 0.089 0.09 0.09 0.09 -1.1 -2.3 -2.3 -2.3
Wheel 3 0.216 0.215 0.216 0.216 0.216 0.5 0.0 0.0 0.0
Wheel 4 0.178 0.179 0.179 0.179 0.178 -0.6 -0.6 -0.6 0.0
Wheel 5 0.094 0.098 0.098 0.098 0.095 -4.3 -4.3 -4.3 -1.1
266

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Off
Wheel 1 0.408 0.429 0.426 0.424 0.409 -5.1 -4.4 -3.9 -0.2
Wheel 2 0.378 0.391 0.39 0.389 0.383 -3.4 -3.2 -2.9 -1.3
Wheel 3 1.855 1.856 1.857 1.857 1.859 -0.1 -0.1 -0.1 -0.2
Wheel 4 1.845 1.845 1.844 1.844 1.842 0.0 0.1 0.1 0.2
Wheel 5 0.353 0.348 0.349 0.349 0.345 1.4 1.1 1.1 2.3

Suspension MR Off
index -3 (x10 ) -3 -3 -3 MR Off
Wheel 1 1.218 1.223 1.209 1.205 1.192 -0.4 0.7 1.1 2.1
Wheel 2 1.181 1.185 1.181 1.178 1.174 -0.3 0.0 0.3 0.6
Wheel 3 14.02 14.081 14.069 14.072 14.068 -0.4 -0.3 -0.4 -0.3
Wheel 4 16.127 16.008 16.023 16.03 16.038 0.7 0.6 0.6 0.6
Wheel 5 1.126 1.133 1.127 1.121 1.11 -0.6 -0.1 0.4 1.4
267
8.5 Summary
In this chapter, the semi-active vibration control of the tracked vehicle suspension with MR
dampers was investigated using a computer simulation model. A mathematical model of 7-DOF
suspension system with three MR dampers was developed and its differential equations were
derived. The Heaviside step function was used as a damper controller while skyhook , hybrid and
fuzz-hybrid were chosen to be system controllers. The proposed semi-active suspension was
compared to a conventional passive suspension for different road profiles and vehicle speeds.
Assessment of the suspension performance was performed in the time and frequency domains
using bounce and pitch accelerations of the body as ride performance criteria. In addition to ride
comfort, the wheel bounce displacements were used as vehicle handling criteria. The simulated
results show that the semi-active control of tracked vehicle suspension integrated with MR
dampers offers a substantial improvement in tracked vehicle ride comfort and handling
characteristics. The suitable selection of fuzzy-hybrid control with a simple Heaviside step function
damper controller will make the application of MR dampers in tracked vehicle suspension more
successful through the alteration of damping forces according to the operating conditions.
268
9 CHAPTER 9: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
This chapter summarizes the outcomes of the research work presented in this thesis, highlighting
important findings. It also presents recommendations for future research to extend this study.
9.1 Conclusions
This dissertation started by considering two key aims. The first aim was to develop a semi-active
suspension system that employs MR dampers to improve the ride comfort and handling
characteristics of tracked vehicles such as the Armour Personnel Carrier (APC) vehicles. The
second aim was to develop various control strategies that enhance the MR dynamic behaviour and
minimize the effect of road-induced vibrations on vehicle dynamic performance over various
terrains. These aims were achieved via the execution of five main objectives. The first main
objective on the theoretical evaluation of the dynamic performance of a tracked vehicle passive
suspension system was achieved in chapters 3 and 4. The second main objective on experimental
charaterisation of the dynamic properties of hydraulic, hydro-gas and MR dampers was achieved in
chapter 5. Main objectives 3 and 4, on design and development of a scaled model suspension
system and on the use of the system for experimental semi-active vibration control tests were
achived in chapters 6 and 7, respectively. The fifth main objective on the evaluation of an intelligent
control method for the semi-active suspension of a full size tracked vehicle was achived in chaper
8. Thus, all the main objectives of the project were achieved.
The main contributions of this study include the following:
1- A conventional passive tracked vehicle suspension system was analysed theoretically. The
Simulink model developed has been validated using published experimental data of a tracked
vehicle field test. The influence of various suspension design parameters such as number
and locations of dampers, damping coefficient, suspension and wheel stiffness on the
suspension dynamic performance were studied. The results proved that the suspension
characteristics have a significant effect on the vehicle dynamic behaviour. The optimal
suspension settings (numbers and locations of the hydraulic dampers) of a full size tracked
vehicle under different excitations have been identified. The suspension fitted with two or
three dampers mounted at the first and last wheels or first, second and last wheels
respectively offer better vehicle performance over bump and sinusoidal terrains. In addition,
the fixed characteristics of the passive damper limit the suspension performance and
decrease the vehicle maximum speeds over rough terrains. Furthermore, the suspension with
low stiffness offers lower acceleration responses than the stiffer suspension system. Finally,
the wheel stiffness affects the suspension dynamic performance. As the wheel stiffness is
increased, the suspension ride performance is degraded.
2- The dynamic characteristics of the hydraulic, hydro-gas and MR dampers have been
evaluated through experimental measurements. These characteristics have been presented
in terms of the force-velocity and force-displacement diagrams. The restoring force surface
269
(RFS) technique has been used to represent the dynamic performance of the three dampers
using a validated Chebyshev polynomial approximation. The resuts show that MR dampers
have more controllable and highly nonlinear characteristics than hydraulic and hydro-gas
dampers. In addition, a modified Bouc-Wen model of the MR damper has been presented
and its predicted response has been validated using different sets of parameters against the
measured dynamic properties of the MR damper. The results showed that the accuracy of the
MR damper model depends on the fitted model parameters. Furthermore, the best set of
parameters to use in the modified Bouc-Wen model has beed derived.
3- A new and novel test rig was designed and fabricated to carry out real-time control of the
scaled tracked vehicle suspension system incorporating MR dampers. The LabVIEW
software and NI-CompactRio hardware were employed for the semi-active control of the
suspension system. The optimal suspension settings i.e. number and locations of the MR
dampers have been identified. The results show that mounting two or three dampers at
extreme wheel locations offer the best suspension performance under various excitations.
This validates the results obtained from the simulation of the full scale tracked vehicle
suspension.
4- The effect of open loop control of the MR damper on the suspension performance was
experimentally evaluated. The measured results show that in open loop control mode of the
MR damper, applications of different constant levels of current to the damper produce
corresponding constant levels of damping force such that at low constant current levels, low
levels of constant damping force are produced and at high constant current levels, high levels
of constant damping force are produced. This limits the suspension capabilities and prevents
the suspension from adapting to the operating conditions.
5- Two semi-active closed loop control methods based on skyhook and hybrid controls were
developed and implemented using the NI-CompactRio data acquisition system to control the
performance of the scaled suspension system fitted with one and two MR dampers. The
influences of the controller gain and weighting factor on the performance of the suspension
with one MR damper have been studied. The skyhook results show that the increase in
controller gain improves the body performance at the expense of the increase in the wheel
response. The hybrid control results prove that the performance of the suspension with the
hybrid control is better than the performance of the suspension with skyhook control. Also,
the weighting factor plays a crucial role in the performance of the hybrid control policy. In
addition, the real time control of the test rig suspension with two MR dampers using both the
skyhook and hybrid control has been presented. The results show that the performance of
passive suspension with two MR dampers located at wheels 1 and 5 is better than the
performance of the suspension with one passive MR damper. Furthermore, the performance
of the suspension is much more improved by using the hybrid control with two MR dampers.
Finally, the measured performance of the undamped and controlled MR suspensions has
been validated against the theoretical Simulink model under sinusoidal excitation.
270
6- A semi-active tracked vehicle suspension model has been developed . This includes the
derivation ofit’s the differential equations that characterise its behaviour. The model has been
validated using published data and it has been used to represent 7-DOF suspension system
of a real tracked vehicle suspension with MR dampers. In this model, three MR dampers
replaced the conventional hydraulic absorbers in the existing conventional passive
suspension. In addition, a modified Bouc-Wen model of a large scale MR damper has been
used to simulate the characteristics of the MR damper for a full size tracked vehicle
application.
7- Three semi-active control strategies based on skyhook, hybrid and fuzzy-hybrid control have
been developed for a full size tracked vehicle suspension. The numerical simulation of the
semi-active suspension with three MR dampers has been performed to demonstrate the
effectiveness of the proposed control methods in reducing the suspension vibration. The
bounce and pitch accelerations of the body were used as a ride performance criteria. In
addition, the bounce displacement of the road wheels was used as a handling performance
indicator. The dynamic behaviour of the passive and controlled tracked vehicle suspension
was assessed using these criteria under bump, sinusoidal and random road excitations. The
results show that the semi-actively controlled suspensions have a better dynamic
performance than the conventional passive suspension. Also, the fuzzy-hybrid control
simultaneously reduces the the vehicle body and the road wheel responses more effectively
than the skyhook and hybrid controls. This results in improvements of the ride comfort and
handling characteristics.
Furthermore, this research has advanced the state of knowledge in semi-active suspension
vibration control. New contributions have been made in semi-active vibration control and non-linear
dynamics of devices such as MR dampers. In addition, this research has resulted in the following
academic publications and/or submitted papers:
1- Ata, W. G. and Oyadiji, S. O., ''Fuzzy hybrid Control of Off-Road Vehicle Suspension Fitted
With Magnetorheological Dampers’’, Proceedings of the ASME 2012, 11th Biennial
Conference on Engineering Systems Design and Analysis; 02-04 Jul 2012; Nantes France.
2012.
2- Ata, W. G. and Oyadiji, S. O., "An investigation into the effect of suspension
configurations on the performance of tracked vehicles traversing bump terrains,"
Vehicle System Dynamics, pp. 1-23, 2014.
3- Ata, W. G. and Oyadiji, S. O., ''An Investigation into the Effect of Suspension
Configurations on the Performance of Tracked Vehicles over Sinusoidal Terrains'',
(submitted to Multibody System Dynamics journal; under review).
271
4- Ata, W. G. and Oyadiji, S. O., ''Semi-Active Vibration Control for Tracked Vehicle
Suspension Systems Incorporating Magnetorheological Dampers'', (submitted to Journal
of Intelligent Material Systems and Structures; under review).
9.2 Recommendations for future work

Although the objectives of this study have been completed, some topics have been identified for
future research. These include the following ideas:
1- Design and development of a new large scale semi-active MR damper with the specific
requirement for tracked vehicle applications.
2- Experimental characterisation of the dynamic properties of the large scale MR damper.

This can help the development of new MR damper model that closely capture the non-
linear dynamic characteristics.
3- Theoretical and experimental characterisation of the friction between the carriages and the
rails of the experimental test rig developed, development of a friction model, inclusion of
the friction model in the 7-DOF suspension systems model, and use of the enhanced
model to predict the characteristics of the test rig.
4- Design and development of a robotic scaled model of a tracked vehicle incorporating MR

dampers and microcontrollers for the proposed control methods. Experimental evaluation
of the ride and handling of the vehicle on bump, sinusoidal and random road profiles and
comparisons of the performance with the predictions of the 7-DOF suspension systems
model. This will provide valuable data for the improvement of the designs of full size
tracked vehicles.
5- Real-time testing of large scale MR dampers and control methods on an actual tracked
vehicle is recommended. Comparison between the real-time testing and the results
obtained from simulation would be useful as well.
272
A REFERENCES
[1] Yagiz, N., Sakman, L. E., and Guclu, R., "Different control applications on a vehicle using
fuzzy logic control," Sadhana - Academy Proceedings in Engineering Sciences, vol. 33, pp.
15-25, 2008.
[2] Eslaminasab, N., "Development of a Semi-active Intelligent Suspension System for Heavy
Vehicles," PhD., mechanical Engineering, Waterloo, Waterloo, Ontario, Canada, 2008.
[3] Biglarbegian, M., Melek, W., and Golnaraghi, F., "Intelligent control of vehicle semi-active
suspension systems for improved ride comfort and road handling," 2006, pp. 19-24.
[4] Baillie, A. S., "Development of a Fuzzy Logic Controller for an Active Suspension of an Off-
road Vehicle Fitted with Terrain Preview," PhD Thesis, 2000.
[5] Karnopp, D. C. and Crosby, M. J., "System for Controlling the Transmission of Energy
Between Spaced Members," USA Patent, 1974.
[6] BenLahcene, Z., Faris, W. F., Khan, M., and Ihsan, S. I., "Semi-active suspension system
for off-road vehicles," in Proceedings of the 3rd International Conference on Mechatronics,
I OM’ 8 8 – 20 December 2008, Kuala Lumpur, Malaysia, 2008.
[7] Lingjie, K., "Study on semi-active suspension system of tracked vehicle based on variable
universe fuzzy control," in 2011 IEEE International Conference on Mechatronics and
Automation (ICMA 2011), Beijing, China, 2011, pp. 2276-80.
[8] Park, J. H. and Park, O. O., "Electrorheology and magnetorheology," Korea-Australia

Rheology Journal, vol. 13, pp. 13-17, 2001.
[9] Rogers, C. A., "Intelligent material systems—the dawn of a new materials age," Journal of
Intelligent Material Systems and Structures, vol. 4, pp. 4-12, 1993.
[10] Butz, T. and von Stryk, O., "Modelling and Simulation of Electro- and Magnetorheological
Fluid Dampers," ZAMM - Journal of Applied Mathematics and Mechanics / Zeitschrift für
Angewandte Mathematik und Mechanik, vol. 82, pp. 3-20, 2002.
[11] Kasim, S. Y., "Ride Analysis For Suspension System of Off-Road Tracked Vehicles," PhD
Thesis, Mechanics, Cranfield Institute of Technology, 1990-91.
[12] Goncalves, F. D., "Dynamic Analysis of Semi-Active Control Techniques of Vehicle

Applications," Msc, Mechanical Engineering, Virginia Polytechnic Institute and State
University, 2001.
[13] Army, U. S., Operator's Manual Carrier, Personnel, Full tracked, Armored, M113 A2, 1984.
[14] Zhang, M. and Zhang, Y., "Fuzzy Control of Semi-Active Suspension Based on 7-DOF
Tracked Vehicle," Journal of Beijing Institute of Technology, vol. 17, 2008.
[15] U.S.Army, "Principles of Automotive Vehicles," Headquarters DEpartment of the Army,

Washington, Technical Manual1995.
[16] Yeh, E. C. and Tsao, Y. J., "Fuzzy control for active suspension design," in In Proceedings
of the Intelligent Vehicle Symposium, Detroit, Michigan, 1992, pp. 92–97, 1992, pp. 92-97.
[17] Faris, W. F., BenLahcene, Z., and Ihsan, S. I., "Assessment of different semi-active control
strategies on the performance of off-road vehicle suspension systems," International
Journal of Vehicle Systems Modelling and Testing, vol. 5, pp. 254-271, 2010.
[18] Lam, A. H. F. and Liao, W. H., "Semi-active control of automotive suspension systems with
magneto-rheological dampers," International Journal of Vehicle Design, vol. 33, pp. 50-75,
2003.
273
[19] Zeng, Y. H., Liu, S. J., and Cheng, W., "Study on Intelligent Control Strategy for Semi-
Active Suspension System of Tracked Vehicle," Applied Mechanics and Materials, vol. 48,
pp. 1162-1171, 2011.
[20] Lee, K., "Numerical modelling for the hydraulic performance prediction of automotive
monotube dampers," Vehicle System Dynamics, vol. 28, pp. 25-39, 1997.
[21] Allen, P., "Models for the dynamic simulation of tank track components," 2006.
[22] Krueger, W. R., Vaculin, O., and Spieck, M., "Evaluation of Active Damping for Reduction
of Noise, Vibration and Motion of Ground Vehicles by Multibody Simulation," DTIC
Document2004.
[23] G.SRINIVASAN, BASHA, A. M. J., and KUMAR, M. S., "Design Studies and Mathematical
Modeling of Hydro-Pneumatic Suspension for Armoured Fighting Vehicles," presented at
the The 11th Asian International Conference on Fluid Machinery and The 3rd Fluid Power
Technology, 2011.
[24] Maclaurin, B., "Progress in British tracked vehicle suspension systems," 1983.
[25] Solomon, U. and Padmanabhan, C., "Hydro-gas suspension system for a tracked vehicle:
Modeling and analysis," Journal of Terramechanics, vol. 48, pp. 125-137, 2011.
[26] Block, H., Kelly, and JP, "Electro-rheology," Journal of Physics D: Applied Physics, vol. 21,
p. 1661, 1988.
[27] Ehrgott, R. and Masri, S., "Modeling the oscillatory dynamic behaviour of electrorheological
materials in shear," Smart Materials and Structures, vol. 1, p. 275, 1992.
[28] Stangroom and J.E., "Electrorheological fluids," Physics in Technology, vol. 14, p. 290,
1983.
[29] Kamath, G. M. and Wereley, N. M., "A nonlinear viscoelastic-plastic model for
electrorheological fluids," Smart Materials and Structures, vol. 6, p. 351, 1997.
[30] Ribakov, Y. and Gluck, J., "Active control of MDOF structures with supplemental
electrorheological fluid dampers," Earthquake Engineering and Structural Dynamics, vol.
28, pp. 143-156, 1999.
[31] Jolly, M. R., Bender, J. W., and Carlson, J. D., "Properties and applications of commercial
magnetorheological fluids," Journal of Intelligent Material Systems and Structures, vol. 10,
pp. 5-13, 1999.
[32] Carlson, J., Catanzarite, D., and Clair, K. A. S., "Commercial magnetorheological fluid
devices," International Journal of Modern Physics B, vol. 10, pp. 2857-2866, 1996.
[33] Dyke, S. J., Bf Jr, S., Sain, M. K., and Carlson, J. D., "Modeling and control of
magnetorheological dampers for seismic response reduction," Smart Materials and
Structures, vol. 5, p. 565, 1996.
[34] Karkoub M.A., Z. M., "Active/semi-active suspension control using magnetorheological

actuators," International journal of systems science, vol. 37, pp. 35-44, 2006.
[35] Yao, G., Yap, F., Chen, G., Li, W. H., and Yeo, S., "MR damper and its application for
semi-active control of vehicle suspension system," Mechatronics, vol. 12, pp. 963-973,
2002.
[36] Koo, J. H., Goncalves, F. D., and Ahmadian, M., "A comprehensive analysis of the
response time of MR dampers," Smart Materials and Structures, vol. 15, p. 351, 2006.
274
[37] Şahin, I., ngin, T., and Çeşmeci, Ş., " omparison of some existing parametric models for
magnetorheological fluid dampers," Smart Materials and Structures, vol. 19, p. 035012,
2010.
[38] Or, S. W., Duan, Y. F., Ni, Y. Q., Chen, Z. H., and Lam, K. H., "Development of
magnetorheological dampers with embedded piezoelectric force sensors for structural
vibration control," Journal of Intelligent Material Systems and Structures, vol. 19, pp. 1327-
1338, 2008.
[39] Ashfak, A., Saheed, A., Rasheed, A., K. K., Jaleel, and A., J., "Design, Fabrication and
Evaluation of MR Damper."
[40] Koo, J. H., "Using magneto-rheological dampers in semiactive tuned vibration absorbers to
control structural vibrations," Virginia Polytechnic Institute and State University, 2003.
[41] Choi, B., S., Lee, K., S., Park, and P., Y., "A hysteresis model for the field dependent
damping force of a magnetorheological damper," Journal of Sound and Vibration, vol. 245,
pp. 375-383, 2001.
[42] Spencer Jr, B. F. D., S. J. Sain, M. K. Carlson, J. D., "Phenomenological model for
magnetorheological dampers," Journal of Engineering Mechanics, vol. 123, pp. 230-238,
1997.
[43] Song, X., Ahmadian, M., and Southward, S. C., "Modeling magnetorheological dampers
with application of nonparametric approach," Journal of Intelligent Material Systems and
[44] Stanway, R., Sproston, J., and Stevens, N., "Non-linear modelling of an electro-rheological
vibration damper," Journal of Electrostatics, vol. 20, pp. 167-184, 1987.
[45] Shen, Y., Golnaraghi, M., and Heppler, G., "Analytical and experimental study of the
response of a suspension system with a magnetorheological damper," Journal of Intelligent
Material Systems and Structures, vol. 16, p. 135, 2005.
[46] Hudha, K., Jamaluddin, H., Samin, P. M., and Rahman, R. A., "Non-parametric linearised
data driven modelling and force tracking control of a magnetorheological damper,"
International Journal of Vehicle Design, vol. 46, pp. 250-269, 2008.
[47] Liao, W. and Wang, D., "Semiactive vibration control of train suspension systems via
magnetorheological dampers," Journal of Intelligent Material Systems and Structures, vol.
14, p. 161, 2003.
[48] Xia, P. Q., "An inverse model of MR damper using optimal neural network and system
identification," Journal of Sound and Vibration, vol. 266, pp. 1009-1023, 2003.
[49] Wang, D. and Liao, W., "Modeling and control of magnetorheological fluid dampers using
neural networks," Smart Materials and Structures, vol. 14, p. 111, 2005.
[50] Metered, H., Bonello, P., and Oyadiji, S. O., "The experimental identification of
magnetorheological dampers and evaluation of their controllers," Mechanical Systems and
Signal Processing, vol. 24, pp. 976-994, 2010.
[51] Metered, H., Bonello, P., and Oyadiji, S. O., "An investigation into the use of neural
networks for the semi-active control of a magnetorheologically damped vehicle
suspension," Proceedings of the Institution of Mechanical Engineers, Part D: Journal of
Automobile Engineering, vol. 224, pp. 829-848, 2010.
[52] Metered, H., Bonello, P., and Oyadiji, S., "Nonparametric identification modeling of
magnetorheological damper using Chebyshev polynomials fits," SAE International Journal
of Passenger Cars-Mechanical Systems, vol. 2, pp. 1125-1135, 2009.
275
[53] Stanway, R., Sproston, L., J., Stevens, and G., N., "Non-linear modelling of an electro-
rheological vibration damper," Journal of Electrostatics, vol. 20, pp. 167-184, 1987.
[54] Hudha, K., Jamaluddin, H., Samin, P., and Rahman, R., "Effects of control techniques and
damper constraint on the performance of a semi-active magnetorheological damper,"
International journal of vehicle autonomous systems, vol. 3, pp. 230-252, 2005.
[55] Rodrigues, F. A., Thouverez, F., Gibert, C., and Jezequel, L., "Chebyshev Polynomials Fits
for Efficient Analysis of Finite Length Squeeze Film Damped Rotors," Journal of
Engineering for Gas Turbines and Power, vol. 125, pp. 175-183, 2003.
[56] Neelakantan, V. A. and Washington, G. N., "Vibration Control of Structural Systems using
MR dampers and a Modified Sliding Mode Control Technique," Journal of Intelligent
Material Systems and Structures, vol. 19, p. 211, 2008.
[57] Wang, D. and Liao, W., "Semiactive controllers for magnetorheological fluid dampers,"
Journal of Intelligent Material Systems and Structures, vol. 16, p. 983, 2005.
[58] Lai, C. and Liao, W., "Vibration control of a suspension system via a magnetorheological
fluid damper," Journal of Vibration and Control, vol. 8, p. 527, 2002.
[59] Du, H., Yim Sze, K., and Lam, J., "Semi-active H-infinity control of vehicle suspension with
magneto-rheological dampers," Journal of Sound and Vibration, vol. 283, pp. 981-996,
2005.
[60] Zhou, L., "Neural network emulation of inverse dynamics for a magnetorheological
damper," Journal of structural Engineering, vol. 128, p. 231, 2002.
[61] 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
[62] Hong, K. S., Sohn, H. C., and Hedrick, J. K., "Modified skyhook control of semi-active
suspensions: a new model, gain scheduling, and hardware-in-the-loop tuning," Journal of
dynamic systems, measurement, and control, vol. 124, p. 158, 2002.
[63] Ahmadian, M. and Pare, C. A., "A quarter-car experimental analysis of alternative
semiactive control methods," Journal of Intelligent Material Systems and Structures, vol.
11, p. 604, 2000.
[64] Masi, J. W., "Effect of Control Techniques on the Performance of Semi-active Dampers,"
MSc., Mechanical Engineering, Virginia Polytechnic Institute and State University, 2001.
[65] Choi, S. B., Lee, H. S., and Park, Y. P., "H8 Control Performance of a Full-Vehicle
Suspension Featuring Magnetorheological Dampers," Vehicle System Dynamics, vol. 38,
pp. 341-360, 2002.
[66] Nakagawa, T. and Sagara, N., "A proposal of a nonlinear H control method for a
semiactive damper with a magnetic fluid base," Magnetics, IEEE Transactions on, vol. 33,
pp. 4206-4208, 1997.
[67] Yokoyama, M., Hedrick, J. K., and Toyama, S., "A model following sliding mode controller
for semi-active suspension systems with MR dampers," 2001, pp. 2652-2657 vol. 4.
[68] Choi, S. B. and Han, Y. M., "Vibration control of electrorheological seat suspension with
human-body model using sliding mode control," Journal of Sound and Vibration, vol. 303,
pp. 391-404, 2007.
[69] Deng, Z., Gao, F., and Liu, X., "Fuzzy control on semi-active suspension based on MR
damper," 2006, p. 63584S.
276
[70] Guan, J.-F., Gu, L., Hou, C.-Z., and Wang, G.-L., "Fuzzy logic control for semi-active
suspension system of tracked vehicle," Journal of Beijing Institute of Technology (English
Edition), vol. 13, pp. 113-117, 2004.
[71] Yu, Y., Wei, X.-x., and Zhang, Y.-f., "Fuzzy logic control for suspension systems of tracked
vehicles," Journal of Beijing Institute of Technology, vol. 18, pp. 37-40, 2009.
[72] Choi, S. B., Park, D. W., and Suh, M. S., "Fuzzy sky-ground hook control of a tracked
vehicle featuring semi-active electrorheological suspension units," Journal of Dynamic
Systems, Measurement and Control, Transactions of the ASME, vol. 124, pp. 150-157,
2002.
[73] Yu, M., Liao, C., Chen, W., and Huang, S., "Study on MR semi-active suspension system
and its road testing," Journal of Intelligent Material Systems and Structures, vol. 17, p. 801,
2006.
[74] Kashani, R. and Strelow, J. E., "Fuzzy logic active and semi-active control of off-road
vehicle suspensions," Vehicle System Dynamics, vol. 32, pp. 409-420, 1999.
[75] Craft, M. J., Buckner, G. D., and Anderson, R. D., "Fuzzy logic control algorithms for
MagneShock semiactive vehicle shock absorbers: design and experimental evaluations,"
2003, p. 577.
[76] Cherry, A. and Jones, R., "Fuzzy logic control of an automotive suspension system," in
IEEE Proceeding-Control Theory Appl., Vol. 142, No. 2, March, 1995, pp. 149-160.
[77] Choi, S. B., Suh, M. S., Park, D. W., and Shin, M. J., "Neuro-fuzzy control of a tracked
vehicle featuring semi-active electro-rheological suspension units," Vehicle System
Dynamics, vol. 35, pp. 141-162, 2001.
[78] Zeng, Y., Liu, S., and E, J., "Neuron PI control for semi-active suspension system of
tracked vehicle," Journal of Central South University of Technology, vol. 18, pp. 444-450.
[79] Ha, S., Choi, S., Rhee, E., and Kang, P., "Optimal design of a magnetorheological fluid
suspension for tracked vehicle," in 11th Conference on Electrorheological Fluids and
Magnetorheological Suspensions, Journal of Physics: Conference Series 149, 2009, p.
012053.
[80] Yu, Y., Wei, X., and Zhang, Y., "Analyses and simulation of fuzzy logic control for
suspension system of a track vehicle," Journal of Beijing Institute of Technology, vol. 17,
pp. 164-7, 2008.
[81] Yuchun, Z., Liangxi, W., Lin, Z., and Hua, C., "Robust and fuzzy control of high mobility off-
road tracked vehicle with trailing arm suspension unit," in 5th World Congress on Intelligent
Control and Automation, 15-19 June 2004, Piscataway, NJ, USA, 2004, pp. 535-9.
[82] Dhir, A. and Sankar, S., "Ride Dynamics of High-Speed Tracked Vehicles: Simulation with
Field Validation," Vehicle System Dynamics, vol. 23, pp. 379-409, 1994.
[83] U.S.Army, "Technical Manual Transport Guidance M113 Family of Vehicles," TM 9-2350-
261-10, 1997.
[84] U.S.Army, "Transport Guidance M113 Family of Vehicles," TM 55-2350-224-14, 1993.
[85] Choi, S. B. and Kim, W. K., "Vibration control of a semi-active suspension featuring
electrorheological fluid dampers," Journal of Sound and Vibration, vol. 234, pp. 537-546,
2000.
[86] Dhir, A., "Ride Dynamics of High Mobility Wheeled/Tracked Off-Road Vehicles: Computer
Simulation with Field Validation," PhD Thesis, Concordia University, 1993.
277
[87] Gündoğdu, Ö., "Optimal seat and suspension design for a quarter car with driver model
using genetic algorithms," International Journal of Industrial Ergonomics, vol. 37, pp. 327-
332, 2007.
[88] Theory of Suspension vol. A-573. MTC,Cairo,Egypt, 1974.
[89] Duym, S., "An alternative force state map for shock absorbers," Proceedings of the
Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, vol. 211,
pp. 175-179, 1997.
[90] Oyadiji S. O. and Sarafianos P., "Characterisation and comparison of the dynamic
properties of conventional and electro-rheological fluid shock absorbers," International
Journal of Vehicle Design, vol. 33, pp. 251-278, 2003.
[91] Yun, H. B., Tasbighoo, F., Masri, S. F., Caffrey, J. P., Wolfe, R. W., Makris, N., and Black,
C., "Comparison of modeling approaches for full-scale nonlinear viscous dampers," Journal
of Vibration and Control, vol. 14, pp. 51-76, 2008.
[92] Worden, K., Hickey, D., Haroon, M., and Adams, D. E., "Nonlinear system identification of
automotive dampers: a time and frequency-domain analysis," Mechanical Systems and
Signal Processing, vol. 23, pp. 104-126, 2009.
[93] Surace, C., Worden, K., and Tomlinson, G., "On the non-linear characteristics of
automotive shock absorbers," Proceedings of the Institution of Mechanical Engineers, Part
D: Journal of Automobile Engineering, vol. 206, pp. 3-16, 1992.
[94] Masri, S. F. and Caughey, T. K., "A nonparametric identification technique for nonlinear
dynamic problems," Journal of Applied Mechanics, vol. 46, p. 433, 1979.
[95] Worden, K., "Parametric and nonparametric identification of nonlinearity in structural

dynamics," PhD, Heriot-Watt University, 1989.
[96] Sarafianos, P. and Oyadiji, S. O., "Characterisation and comparison of the dynamic
properties of conventional and electro-rheological fluid shock absorbers," International
Journal of Vehicle Design, vol. 33, pp. 251-278, 2003.
[97] Chooi, W. W. and Oyadiji, S. O., "Design, modelling and testing of magnetorheological
(MR) dampers using analytical flow solutions," Computers and Structures, vol. 86, pp. 473-
482, 2008.
[98] Spencer, J. r., Dyke, B. F., Sain, S. J., Carlson, M. K., and D., J., "Phenomenological
model for magnetorheological dampers," Journal of Engineering Mechanics, vol. 123, pp.
230-238, 1997.
[99] Wang, D. and Liao, W., "Magnetorheological fluid dampers: a review of parametric
modelling," Smart Materials and Structures, vol. 20, p. 023001, 2011.
[100] Tomlinson, G. and Oyadiji, S. O., "Determination of the complex moduli of viscoelastic
structural elements by resonance and non-resonance methods," Journal of Sound and
Vibration, vol. 101, pp. 277-298, 1985.
[101] Park, D., Choi, S. B., Suh, M. S., and Shin, M., "ER suspension units for vibration control of
a tracked vehicle," in Smart Structures and Materials 2001: Smart Structures and
Integrated Systems, L. Porter Davis,Editor, Proceedings of SPIE Vol. 4327 (2001), 2001, p.
159.
[102] Dhir, A., Sankar, and Seshadri, "Assessment of tracked vehicle suspension system using a
validated computer simulation model," Journal of Terramechanics, vol. 32, pp. 127-149,
1995.
278
Appendix A: Test rig design schematics
This appendix presents the bill of materials for the MR suspension test rig. Also, It includes the
detail drawings for the test rig main parts. Finally, the process of the assembly of the test rig
components is described.
A.1 Bill of materials for the test rig
Many materials were needed for the manufacture of the MR test rig suspension system. Table A.1
summarizes all the materials and components which are used to build the test rig. The table
presents the components, the quantities and the material specifications. Also, the front, top and
side views of the test rig are shown in Figure A.1.
Table A- 1: Bill of Materials for rig suspension system

No Component Material Qt
1 Damper joint Aluminium 10
2 Base plate Aluminium 1
3 Y-bearing stand Steel 6
4 Body mass pivot Steel 1
5 Body mass ball bearing Energy efficient bearing 30mm ID,55mm 2
6 Body mass Aluminium 1
7 Fixing bush steel 10
8 Carriage LLTHC30SU-T0P5-*SKF Linear Carriage 10
9 Body mass carriage LLTHC30A-T0P5-*SKF Linear Carriage 1
10 Connecting rod Aluminium 5
11 Connecting rod joint Aluminium 5
12 Connecting rod lower bearing Plain bearing with flange-GFM6065-22 5
13 Connecting rod upper bearing Plain bearing with flange-GFM1214-20 5
14 MR damper RD-8040-1 5
15 Excentric cam Aluminium 5
16 Excitation mass (lower part) Aluminium 5
17 Excitation mass (upper part) Aluminium 5
18 Force sensor 44kN ring force sensor - 4.45 pC/N (1210C2) 5
19 Ground plate Steel 1
20 Cross beamr Aluminium 1
21 Wheel spring Compression spring (code: D23560) 5
22 Suspension spring clamp (lower part) Aluminium 5
th
23 Wheel mass (4 wheel) Aluminium 1
24 Geared motor Tec 4kw 4pole 3phase d112m 1
25 Split cotter pin P1240.05-40-mild steel 1
26 Clevis pin (MR damper) P1242.12-120 5
27 Clevis pin (connecting rod) P1242.12-45 5
28 Rail Llthr30p5rail-*skf linear rail 5
29 Crank shaft Steel 1
30 Carriage spacer Aluminium 10
31 Suspension spring Compression spring (code:227) 5
32 wheel spring clamp (lower part) Aluminium 5
33 Suspension spring pillow Aluminium 5
34 Stud bolt Steel 6
35 Suspension spring clamp (upper part) Aluminium 5
36 Side beam Aluminium 5
37 Washer Steel 10
nd
38 Wheel mass (2 wheel) Aluminium 1
39 Crank shaft bearing Set screw ball bearing YAR 206-2RF 6
279
Figure A.1 Front, top and side view of the test rig
280
A.2 Detail drawings
This section presents the detail drawings for the main components of the test rig.
1- Body mass
281
2- Wheel mass
282
3- Excitation mass (upper)
283
4- Excitation mass (lower)
284
5- Crank shaft
285
6- Eccentric cam
286
7- Connecting rod
287
8- Base plate
288
9- Side beam
289
10- Cross beam
290
A.3 Assembly process
1- Frame assembly
Figure A.2 shows the test rig frame which consists of five side beams and one cross beam. The
side beams are linked together by the cross beam through cap screw bolts. Then, the frame is
bolted to the base plate (not shown) which is tightened to the ground. This frame is then used to fix
the rails on which ten carriages can slide freely in the vertical plane.
Side beams
Carriages
Rails
Cross beam
Figure A.2 Test rig main frame
2- Drive mechanism assembly

Figure A.3 shows the drive mechanism which consists of the crank shaft, five eccentric cams, five
connecting rods, ten bushes, six bearing stands and six Y-bearings. First, the shaft has five keys
with different orientations. The cams which have eccentric holes with keys can be mounted on the
shaft with the specified directions (phase angle). Then, the connecting rods are fitted to the cams
through connecting rod bearings. The side motion of the connecting rods are prevented using ten
bushes.
Y-Bearings
Conn. rods Cams Bushes
Figure A.3 Drive mechanism components
291
3- Excitation mass assembly
Figure A.4 shows the excitation masses assembly which include two parts (upper and lower).
These two parts are bolted together with four cap screws. The lower parts are connected to the
connecting rods through five joints. In addition, the upper parts are connected to the wheel
compresson springs through spring pillows. Each excitation mass is connected to a small carriage
and can slide freely in the vertical direction on the rails. The excitation masses transmit the
sinusoidal excitations generated from the drive mechanism to the wheel masses.
Wheel springs
Joint
s
Excitation mass
(upper)
Excitation mass
(lower) Carriage
Figure A.4 Excitation masses assembly
4- Wheel mass assembly
The wheel mass assembly is shown in Figure A.5. In the figure, each wheel mass is connected
from the lower side to the upper part of the wheel spring. The upper side has a slot to
accommodate the lower end of the MR damper and a pillow to fix the lower part of the suspension
compression spring. The pillow is tightened with two cap screws to the wheel mass. Also, the pillow
has a cap screw to tight the spring clamp. Furthermore, each wheel mass is attached to a carriage
and the whole part can slide freely on the rails in the vertical direction.
5- Body mass assembly
Figure A.6 shows the body mass assembly. It consists of a rectangular bar attached to a carriage
through a pivot. This allows the body mass to rotate about its centre of gravity and oscillate in the
vertical direction. The body mass is suspended over five compression springs which are tightened
from their upper end with a clamp. Also, the body mass has five cylindrical grooves for mounting
the upper end of the MR dampers and the force sensors.
292
Wheel masses
Clamp
Spring pillow
Figure A.5 Wheel mass assembly
Body mass
Carriage
Force sensors
Pivot
Figure A.6 Body mass assembly
293
Appendix B: Compression spring details
Table B.1 Test rig compression spring details
Parameters Suspension springs Wheel springs
Wire dia. (mm) 4.88 4
Outside dia. (mm) 38.1 24
Free length (mm) 152.4 105
Approx. No. of coils 15 14.5
Rate (N/mm) 11.73 21.73
Code No. 227 D23560
Company name http://www.entexstocksprings.co.uk http://www.springmasters.com
294

Wael Kotb Ata Thesis Full-Text PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wael Kotb Ata Thesis Full-Text PDF

Uploaded by

Copyright:

Available Formats

Intelligent Control of Tracked

A thesis submitted to the University of Manchester for the degree of

Wael Galal Kotb Ata

School of Mechanical, Aerospace and Civil Engineering

A List of Figures .........................................................................................................................5

1 CHAPTER 1: INTRODUCTION .................................................................................................25

1.1 Background .........................................................................................................................25

2 CHAPTER 2: LITERATURE REVIEW ......................................................................................30

2.1 Introduction .........................................................................................................................30

3 CHAPTER 3: EVALUATION OF TRACKED VEHICLE PASSIVE SUSPENSION

4 CHAPTER 4: EVALUATION OF TRACKED VEHICLE PASSIVE SUSPENSION

4.1 Introduction ....................................................................................................................... 107

5 CHAPTER 5: EXPERIMENTAL IDENTIFICATION OF HYDRAULIC, HYDRO-GAS AND

5.1 Introduction ....................................................................................................................... 152

6 CHAPTER 6: DESIGN, MANUFACTURE AND INSTRUMENTATION OF TRACKED

6.1 Introduction .......................................................................................................................185

7 CHAPTER 7: SEMI-ACTIVE CONTROL OF TEST RIG SUSPENSION WITH

7.1 Introduction .......................................................................................................................197

8 CHAPTER 8: SEMI-ACTIVE CONTROL OF TRACKED VEHICLE SUSPENSION WITH MR

8.1 Introduction ....................................................................................................................... 234

9 CHAPTER 9: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK ........... 269

9.1 Conclusions ...................................................................................................................... 269

APPENDIX A: TEST RIG DESIGN SCHEMATICS ....................................................................... 279

APPENDIX B: COMPRESSION SPRING DETAILS ..................................................................... 294

Figure ‎2.1 Different types of tracked vehicles suspension system ...................................................30

Figure ‎2.5 Sectional view of the rotary-vane damper [21] ................................................................35

Figure ‎2.7 Magnetic field effect on MR fluid [33]...............................................................................36

Figure ‎2.8 Sectional view of MR Damper [22] ..................................................................................38

Figure ‎2.9 Mono-tube MR Damper [40] ............................................................................................38

Figure ‎2.10 Twin tube MR Damper [40] ............................................................................................39

Figure ‎2.11 Double-ended MR Damper [40] .....................................................................................39

Figure ‎2.12 Schematic diagrams for MR damper parametric models ..............................................40

Figure ‎2.13 Idealized skyhook control schemes [63] ........................................................................44

Figure ‎3.1 Half tracked vehicle suspension model ...........................................................................48

Figure ‎3.2 Typical bump road profiles at vehicle speed of 10 km/h..................................................50

Figure ‎3.3 Top-level diagram of the passive suspension model .......................................................51

Figure ‎3.23 Influence of suspension stiffness on BA responses under medium bump.................... 93

Figure ‎3.24 Influence of suspension stiffness on AA responses under medium bump.................... 94

‎ .8 Influence of varying damping coefficient on BA responses under shallow sinusoidal road

‎ .10 Influence of varying damping coefficient on BA responses under medium sinusoidal

‎ .11 Influence of varying damping coefficient on AA responses under medium sinusoidal

‎ .14 Influence of varying suspension stiffness on BA responses under shallow sinusoidal

‎ .15 Influence of varying suspension stiffness on AA responses under shallow sinusoidal

‎ .16 Influence of varying suspension stiffness on BA responses under medium sinusoidal

‎ .17 Influence of varying suspension stiffness on AA responses under medium sinusoidal

‎ .3 Force-displacement and force-velocity dependence of the hydraulic damper under

‎ .4 Force-displacement and force-velocity dependence of the hydraulic damper under

‎ .5 Force-displacement and force-velocity dependence of the hydraulic damper under

‎ .6 Force-velocity characteristics for hydraulic damper at 10 mm amplitude and different

‎ .8 Force-displacement and force-velocity dependence of the hydro-gas damper under

‎ .9 Force-displacement and force-velocity dependence of the hydro-gas damper under

‎ .10 Force-displacement and force-velocity dependence of the hydro-gas damper under

‎ .11 Force-velocity characteristics for hydro-gas damper at 10 mm amplitude and different

‎ .13 Measured characteristics of MR damper at 1 Hz excitation frequency and with

‎ .14 Measured characteristics of MR damper at 6 Hz excitation frequency and with

‎ .15 Measured characteristics of MR damper at 11 Hz excitation frequency and with

‎ .16 Measured characteristic of MR damper under different excitation frequencies,

‎ .18 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 2

‎ .20 Comparison of restoring force surface characteristics of dampers under excitation

‎ .21 Comparison of measured characteristics of hydraulic, hydro-gas and MR dampers at 6

‎ .22 Validation of experimental tests and Chebyshev approximation under excitation

‎ .23 Comparison of restoring force surface characteristics of damper under excitation

‎ .26 Comparison between force-displacement and force-velocity characteristics predicted