Vehicle Dynamics

V
E
H
I
C
L
E

D
Y
N
A
M
I
C
S
FACHHOCHSCHULE REGENSBURG
UNIVERSITY OF APPLIED SCIENCES
HOCHSCHULE FR
TECHNIK
WIRTSCHAFT
SOZIALES
LECTURE NOTES
Prof. Dr. Georg Rill
October 2006
download: http://homepages.fh-regensburg.de/%7Erig39165/
Contents
Contents I
1 Introduction 1
1.1 Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Vehicle Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Driver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.3 Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.4 Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.5 Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Denitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.1 Reference frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.2 Toe-in, Toe-out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Wheel Camber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.4 Design Position of Wheel Rotation Axis . . . . . . . . . . . . . . . . . 5
1.2.5 Steering Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.5.1 Kingpin . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.5.2 Caster and Kingpin Angle . . . . . . . . . . . . . . . . . . . 8
1.2.5.3 Caster, Steering Offset and Disturbing Force Lever . . . . . . 8
2 Road 10
2.1 Modeling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Deterministic Proles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.1 Bumps and Potholes . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.2 Sine Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Random Proles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.1 Statistical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.2 Classication of Random Road Proles . . . . . . . . . . . . . . . . . 15
2.3.3 Realizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3.3.1 Sinusoidal Approximation . . . . . . . . . . . . . . . . . . . 16
2.3.3.2 Shaping Filter . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3.3.3 Two-Dimensional Model . . . . . . . . . . . . . . . . . . . 18
3 Tire 19
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.1 Tire Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.2 Tire Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
I
Contents
3.1.3 Tire Forces and Torques . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.1.4 Measuring Tire Forces and Torques . . . . . . . . . . . . . . . . . . . 21
3.1.5 Modeling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Contact Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.1 Basic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.2 Tire Deection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.2.3 Length of Contact Patch . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.4 Static Contact Point . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2.5 Contact Point Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2.6 Dynamic Rolling Radius . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Forces and Torques caused by Pressure Distribution . . . . . . . . . . . . . . . 32
3.3.1 Wheel Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.2 Tipping Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3.3 Rolling Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4 Friction Forces and Torques . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.1 Longitudinal Force and Longitudinal Slip . . . . . . . . . . . . . . . . 35
3.4.2 Lateral Slip, Lateral Force and Self Aligning Torque . . . . . . . . . . 38
3.4.3 Wheel Load Inuence . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.4.4 Different Friction Coefcients . . . . . . . . . . . . . . . . . . . . . . 40
3.4.5 Typical Tire Characteristics . . . . . . . . . . . . . . . . . . . . . . . 41
3.4.6 Combined Slip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4.7 Camber Inuence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.8 Bore Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.4.8.1 Modeling Aspects . . . . . . . . . . . . . . . . . . . . . . . 46
3.4.8.2 Maximum Torque . . . . . . . . . . . . . . . . . . . . . . . 47
3.4.8.3 Bore Slip . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.4.8.4 Model Realisation . . . . . . . . . . . . . . . . . . . . . . . 48
3.5 First Order Tire Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4 Suspension System 50
4.1 Purpose and Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.2 Some Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2.1 Multi Purpose Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2.2 Specic Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.3 Steering Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.3.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.3.2 Rack and Pinion Steering . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.3.3 Lever Arm Steering System . . . . . . . . . . . . . . . . . . . . . . . 53
4.3.4 Drag Link Steering System . . . . . . . . . . . . . . . . . . . . . . . . 54
4.3.5 Bus Steer System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.4 Standard Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.4.1 Springs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.4.2 Anti-Roll Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.4.3 Damper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
II
Contents
4.4.4 Rubber Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.5 Dynamic Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.5.1 Testing and Evaluating Procedures . . . . . . . . . . . . . . . . . . . . 60
4.5.2 Simple Spring Damper Combination . . . . . . . . . . . . . . . . . . . 63
4.5.3 General Dynamic Force Model . . . . . . . . . . . . . . . . . . . . . . 65
4.5.3.1 Hydro-Mount . . . . . . . . . . . . . . . . . . . . . . . . . 66
5 Vertical Dynamics 70
5.1 Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.2 Basic Tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.2.1 From complex to simple models . . . . . . . . . . . . . . . . . . . . . 70
5.2.2 Natural Frequency and Damping Rate . . . . . . . . . . . . . . . . . . 73
5.2.3 Spring Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.3.1 Minimum Spring Rates . . . . . . . . . . . . . . . . . . . . 75
5.2.3.2 Nonlinear Springs . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.4 Inuence of Damping . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.2.5 Optimal Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.2.5.1 Avoiding Overshoots . . . . . . . . . . . . . . . . . . . . . 79
5.2.5.2 Disturbance Reaction Problem . . . . . . . . . . . . . . . . 80
5.3 Sky Hook Damper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3.1 Modelling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3.2 Eigenfrequencies and Damping Ratios . . . . . . . . . . . . . . . . . . 86
5.3.3 Technical Realization . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.4 Nonlinear Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.4.1 Quarter Car Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.4.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6 Longitudinal Dynamics 92
6.1 Dynamic Wheel Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.1.1 Simple Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.1.2 Inuence of Grade . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.1.3 Aerodynamic Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.2 Maximum Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.2.1 Tilting Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.2.2 Friction Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.3 Driving and Braking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.3.1 Single Axle Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.3.2 Braking at Single Axle . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.3.3 Braking Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.4 Optimal Distribution of Drive and Brake Forces . . . . . . . . . . . . . 99
6.3.5 Different Distributions of Brake Forces . . . . . . . . . . . . . . . . . 101
6.3.6 Anti-Lock-Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.4 Drive and Brake Pitch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.4.1 Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
III
Contents
6.4.2 Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.4.3 Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.4.4 Driving and Braking . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.4.5 Brake Pitch Pole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
7 Lateral Dynamics 108
7.1 Kinematic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.1.1 Kinematic Tire Model . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.1.2 Ackermann Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.1.3 Space Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.4 Vehicle Model with Trailer . . . . . . . . . . . . . . . . . . . . . . . . 111
7.1.4.1 Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
7.1.4.2 Vehicle Motion . . . . . . . . . . . . . . . . . . . . . . . . 112
7.1.4.3 Entering a Curve . . . . . . . . . . . . . . . . . . . . . . . . 113
7.1.4.4 Trailer Motions . . . . . . . . . . . . . . . . . . . . . . . . 114
7.1.4.5 Course Calculations . . . . . . . . . . . . . . . . . . . . . . 115
7.2 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.2.1 Cornering Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.2.2 Overturning Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.2.3 Roll Support and Camber Compensation . . . . . . . . . . . . . . . . 120
7.2.4 Roll Center and Roll Axis . . . . . . . . . . . . . . . . . . . . . . . . 123
7.2.5 Wheel Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.3 Simple Handling Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
7.3.1 Modeling Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
7.3.2 Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
7.3.3 Tire Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.4 Lateral Slips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.5 Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.3.6 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.3.6.1 Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.3.6.2 Low Speed Approximation . . . . . . . . . . . . . . . . . . 128
7.3.6.3 High Speed Approximation . . . . . . . . . . . . . . . . . . 128
7.3.6.4 Critical Speed . . . . . . . . . . . . . . . . . . . . . . . . . 129
7.3.7 Steady State Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
7.3.7.1 Steering Tendency . . . . . . . . . . . . . . . . . . . . . . . 130
7.3.7.2 Side Slip Angle . . . . . . . . . . . . . . . . . . . . . . . . 132
7.3.7.3 Slip Angles . . . . . . . . . . . . . . . . . . . . . . . . . . 133
7.3.8 Inuence of Wheel Load on Cornering Stiffness . . . . . . . . . . . . . 134
8 Driving Behavior of Single Vehicles 136
8.1 Standard Driving Maneuvers . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
8.1.1 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . 136
8.1.2 Step Steer Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
8.1.3 Driving Straight Ahead . . . . . . . . . . . . . . . . . . . . . . . . . . 138
IV
Contents
8.1.3.1 Random Road Prole . . . . . . . . . . . . . . . . . . . . . 138
8.1.3.2 Steering Activity . . . . . . . . . . . . . . . . . . . . . . . . 140
8.2 Coach with different Loading Conditions . . . . . . . . . . . . . . . . . . . . 140
8.2.1 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
8.2.2 Roll Steering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
8.2.3 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . 141
8.2.4 Step Steer Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.3 Different Rear Axle Concepts for a Passenger Car . . . . . . . . . . . . . . . . 143
V
1 Introduction
1.1 Terminology
1.1.1 Vehicle Dynamics
Vehicle dynamics is a part of engineering primarily based on classical mechanics but it may
also involve physics, electrical engineering, chemistry, communications, psychology etc. Here,
the focus will be laid on ground vehicles supported by wheels and tires. Vehicle dynamics
encompasses the interaction of:
driver
vehicle
load
environment
Vehicle dynamics mainly deals with:
the improvement of active safety and driving comfort
the reduction of road destruction
In vehicle dynamics are employed:
computer calculations
test rig measurements
eld tests
In the following the interactions between the single systems and the problems with computer
calculations and/or measurements shall be discussed.
1
1 Introduction
1.1.2 Driver
By various means the driver can interfere with the vehicle:
driver
_
_
steering wheel lateral dynamics
accelerator pedal
brake pedal
clutch
gear shift
_
_
longitudinal dynamics
_
_
vehicle
The vehicle provides the driver with these information:
vehicle
_
_
_
vibrations: longitudinal, lateral, vertical
sounds: motor, aerodynamics, tires
instruments: velocity, external temperature, ...
_
_
_
driver
The environment also inuences the driver:
environment
_
_
_
climate
trafc density
track
_
_
_
driver
The drivers reaction is very complex. To achieve objective results, an ideal driver is used in
computer simulations, and in driving experiments automated drivers (e.g. steering machines)
are employed.
Transferring results to normal drivers is often difcult, if eld tests are made with test drivers.
Field tests with normal drivers have to be evaluated statistically. Of course, the drivers security
must have absolute priority in all tests.
Driving simulators provide an excellent means of analyzing the behavior of drivers even in limit
situations without danger.
It has been tried to analyze the interaction between driver and vehicle with complex driver
models for some years.
1.1.3 Vehicle
The following vehicles are listed in the ISO 3833 directive:
motorcycles
passenger cars
busses
trucks
2
1.1 Terminology
agricultural tractors
passenger cars with trailer
truck trailer / semitrailer
road trains
For computer calculations these vehicles have to be depicted in mathematically describable
substitute systems. The generation of the equations of motion, the numeric solution, as well
as the acquisition of data require great expenses. In times of PCs and workstations computing
costs hardly matter anymore.
At an early stage of development, often only prototypes are available for eld and/or laboratory
tests. Results can be falsied by safety devices, e.g. jockey wheels on trucks.
1.1.4 Load
Trucks are conceived for taking up load. Thus, their driving behavior changes.
Load
_
mass, inertia, center of gravity
dynamic behaviour (liquid load)
_
vehicle
In computer calculations problems occur at the determination of the inertias and the modeling
of liquid loads.
Even the loading and unloading process of experimental vehicles takes some effort. When car-
rying out experiments with tank trucks, ammable liquids have to be substituted with water.
Thus, the results achieved cannot be simply transferred to real loads.
1.1.5 Environment
The environment inuences primarily the vehicle:
Environment
_
road: irregularities, coefcient of friction
air: resistance, cross wind
_
vehicle
but also affects the driver:
environment
_
climate
visibility
_
driver
Through the interactions between vehicle and road, roads can quickly be destroyed.
The greatest difculty with eld tests and laboratory experiments is the virtual impossibility of
reproducing environmental inuences.
The main problems with computer simulation are the description of random road irregularities
and the interaction of tires and road as well as the calculation of aerodynamic forces and torques.
3
1 Introduction
1.2 Denitions
1.2.1 Reference frames
A reference frame xed to the vehicle and a ground-xed reference frame are used to describe
the overall motions of the vehicle, Figure 1.1. The ground-xed reference frame with the axis
x
0
y
0
z
0
x
F
y
F
z
F
y
C
z
C
x
C
e
yR
e
n
Figure 1.1: Frames used in vehicle dynamics
x
0
, y
0
, z
0
serves as an inertial reference frame. Within the vehicle-xed reference frame the
x
F
-axis points forward, the y
F
-axis to the left, and the z
F
-axis upward.
The wheel rotates around an axis which is xed to the wheel carrier. The reference frame C is
xed to the wheel carrier. In design position its axes x
C
, y
C
and z
C
are parallel to the corre-
sponding axis of vehicle-xed reference frame F.
The momentary position of the wheel is xed by the wheel center and the orientation of the
wheel rim center plane which is dened by the unit vector e
yR
into the direction of the wheel
rotation axis.
Finally, the normal vector e
n
describes the inclination of the local track plane.
1.2.2 Toe-in, Toe-out
Wheel toe-in is an angle formed by the center line of the wheel and the longitudinal axis of the
vehicle, looking at the vehicle from above, Figure 1.2. When the extensions of the wheel center
lines tend to meet in front of the direction of travel of the vehicle, this is known as toe-in. If,
however the lines tend to meet behind the direction of travel of the vehicle, this is known as
toe-out. The amount of toe can be expressed in degrees as the angle to which the wheels are
out of parallel, or, as the difference between the track widths as measured at the leading and
trailing edges of the tires or wheels.
Toe settings affect three major areas of performance: tire wear, straight-line stability and corner
entry handling characteristics.
4
1.2 Denitions
toe-in toe-out
+
+
y
F
x
F
y
F
x
F
Figure 1.2: Toe-in and Toe-out
For minimum tire wear and power loss, the wheels on a given axle of a car should point directly
ahead when the car is running in a straight line. Excessive toe-in or toe-out causes the tires to
scrub, since they are always turned relative to the direction of travel.
Toe-in improves the directional stability of a car and reduces the tendency of the wheels to
shimmy.
1.2.3 Wheel Camber
Wheel camber is the angle of the wheel relative to vertical, as viewed from the front or the rear
of the car, Fig. 1.3. If the wheel leans away from the car, it has positive camber; if it leans in
+ +
y
F
z
F
e
n

y
F
z
F
e
n
positive camber negative camber
Figure 1.3: Positive camber angle
towards the chassis, it has negative camber. The wheel camber angle must not be mixed up with
the tire camber angle which is dened as the angle between the wheel center plane and the local
track normal e
n
. Excessive camber angles cause a non symmetric tire wear.
A tire can generate the maximum lateral force during cornering if it is operated with a slightly
negative tire camber angle. As the chassis rolls in corner the suspension must be designed such
that the wheels performs camber changes as the suspension moves up and down. An ideal sus-
pension will generate an increasingly negative wheel camber as the suspension deects upward.
1.2.4 Design Position of Wheel Rotation Axis
The unit vector e
yR
describes the wheel rotation axis. Its orientation with respect to the wheel
carrier xed reference frame can be dened by the angles
0
and
0
or
0
and
0
, Fig. 1.4. In
5
1 Introduction
0
e
yR
z
C
= z
F
0
x
C
= x
F
y
C
= y
F
0
*
Figure 1.4: Design position of wheel rotation axis
design position the corresponding axes of the frames C and F are parallel. Then, for the left
wheel we get
e
yR,F
= e
yR,C
=
1
_
tan
2
0
+ 1 + tan
2
0
_
_
tan
0
1
tan
0
_
_
(1.1)
or
e
yR,F
= e
yR,C
=
_
_
sin
0
cos
0
cos
0
cos
0
sin
0
_
_
, (1.2)
where
0
is the angle between the y
F
-axis and the projection line of the wheel rotation axis into
the x
F
- y
F
-plane, the angle
0
describes the angle between the y
F
-axis and the projection line of
the wheel rotation axis into the y
F
- z
F
-plane, whereas
0
0
is the angle between the wheel rotation
axis e
yR
and its projection into the x
F
- y
F
-plane. Kinematics and compliance test machines
usually measure the angle
0
. That is why, the automotive industry mostly uses this angle instead
of
0
.
On a at and horizontal road where the track normal e
n
points into the direction of the vertical
axes z
C
= z
F
the angles
0
and
0
correspond with the toe angle and the camber angle
0
. To
specify the difference between
0
and
0
the ratio between the third and second component of
the unit vector e
yR
is considered. The Equations 1.1 and 1.2 deliver
tan
0
1
=
sin
0
cos
0
cos
0
or tan
0
=
tan
0
cos
0
. (1.3)
Hence, for small angles
0
1 the difference between the angles
0
and
0
is hardly noticeable.
6
1.2 Denitions
1.2.5 Steering Geometry
1.2.5.1 Kingpin
At steered front axles, the McPherson-damper strut axis, the double wishbone axis, and the
multi-link wheel suspension or the enhanced double wishbone axis are mostly used in passenger
cars, Figs. 1.5 and 1.6.
C
A
B
e
S
z
C
x
C
z
C
Figure 1.5: Double wishbone wheel suspension
z
C
y
C
C
x
C
e
S
T
A
rotation axis
z
C
y
C
x
C
e
S
C
Figure 1.6: McPherson and multi-link wheel suspensions
The wheel body rotates around the kingpin line at steering motions. At the double wishbone
axis the ball joints A and B, which determine the kingpin line, are both xed to the wheel
body. Whereas the ball joint A is still xed to the wheel body at the standard McPherson wheel
suspension, the top mount T is now xed to the vehicle body. At a multi-link axle the kingpin
line is no longer dened by real joints. Here, as well as with an enhanced McPherson wheel
suspension, where the A-arm is resolved into two links, the momentary rotation axis serves as
7
1 Introduction
kingpin line. In general the momentary momentary rotation axis is neither xed to the wheel
body nor to the chassis and, it will change its position at wheel travel and steering motions.
1.2.5.2 Caster and Kingpin Angle
The unit vector e
S
describes the direction of the kingpin line. Within the vehicle xed reference
frame F it can be xed by two angles. The caster angle denotes the angle between the z
F
-axis
and the projection line of e
S
into the x
F
-, z
F
-plane. In a similar way the projection of e
S
into
the y
F
-, z
F
-plane results in the kingpin inclination angle , Fig. 1.7.
x
F
y
F
z
F
e
S
z
F
Figure 1.7: Kingpin and caster angle
At many axles the kingpin and caster angle can no longer be determined directly. Here, the
current rotation axis at steering motions, which can be taken from kinematic calculations will
yield a virtual kingpin line. The current values of the caster angle and the kingpin inclination
angle can be calculated from the components of the unit vector e
S
in the direction of the
kingpin line, described in the vehicle xed reference frame
tan =
e
(1)
S,F
e
(3)
S,F
and tan =
e
(2)
S,F
e
(3)
S,F
, (1.4)
where e
(1)
S,F
, e
(2)
S,F
, e
(3)
S,F
are the components of the unit vector e
S,F
expressed in the vehicle xed
reference frame F.
1.2.5.3 Caster, Steering Offset and Disturbing Force Lever
The contact point P, the local track normal e
n
and the unit vectors e
x
and e
y
which point into
the direction of the longitudinal and lateral tire force result from the contact geometry. The axle
kinematics denes the kingpin line. In general, the point S where an extension oft the kingpin
line meets the road surface does not coincide with the contact point P, Fig. 1.8. As both points
are located on the local track plane, for the left wheel the vector from S to P can be written as
r
SP
= c e
x
+ s e
y
, (1.5)
8
1.2 Denitions
S
P
C
d
e
x
e
y
s
c
e
n
kingpin
line
e
S
local track
plane
e
yR
wheel
rotation
axis
Figure 1.8: Caster and Steering offset
where c names the caster and s is the steering offset. Caster and steering offset will be positive,
if S is located in front of and inwards of P.
The distance d between the wheel center C and the king pin line represents the disturbing force
lever. It is an important quantity in evaluating the overall steering behavior, [24].
9
2 Road
2.1 Modeling Aspects
Sophisticated road models provide the road height z
R
and the local friction coefcient
L
at
each point x, y, Fig. 2.1.
z(x,y)
x
0
y
0
z
0
(x,y)
Center Line L(s)
Friction
Segments
Road profile
Obstacle
Figure 2.1: Sophisticated road model
The tire model is then responsible to calculate the local road inclination. By separating the
horizontal course description from the vertical layout and the surface properties of the roadway
almost arbitrary road layouts are possible, [4].
Besides single obstacles or track grooves the irregularities of a road are of stochastic nature. A
vehicle driving over a random road prole mainly performs hub, pitch and roll motions. The
local inclination of the road prole also induces longitudinal and lateral motions as well as yaw
motions. On normal roads the latter motions have less inuence on ride comfort and ride safety.
To limit the effort of the stochastic description usually simpler road models are used.
If the vehicle drives along a given path its momentary position can be described by the path
variable s = s(t). Hence, a fully two-dimensional road model can be reduced to a parallel track
model, Fig. 2.2.
10
2.2 Deterministic Proles
z
1
(s)
s
x
y
z
z
R
(x,y)
z
1
z
2
Figure 2.2: Parallel track road model
Now, the road heights on the left and right track are provided by two one-dimensional functions
z
1
= z
1
(s) and z
2
= z
2
(s). Within the parallel track model no information about the local
lateral road inclination is available. If this information is not provided by additional functions
the impact of a local lateral road inclination to vehicle motions is not taken into account.
For basic studies the irregularities at the left and the right track can considered to be approxi-
mately the same, z
1
(s) z
2
(s). Then, a single track road model with z
R
(s) = z
1
(x) = z
2
(x)
can be used. Now, the roll excitation of the vehicle is neglected too.
2.2 Deterministic Proles
2.2.1 Bumps and Potholes
Bumps and Potholes on the road are single obstacles of nearly arbitrary shape. Already with
simple rectangular cleats the dynamic reaction of a vehicle or a single tire to a sudden impact
can be investigated. If the shape of the obstacle is approximated by a smooth function, like a
cosine wave, then, discontinuities will be avoided. Usually the obstacles are described in local
reference frames, Fig. 2.3.
L
H
B B
x
y
z
H
L
x
y
z
Figure 2.3: Rectangular cleat and cosine-shaped bump
Then, the rectangular cleat is simply dened by
z(x, y) =
_
H if 0 < x < L and
1
2
B < y <
1
2
B
0 else
(2.1)
11
2 Road
and the cosine-shaped bump is given by
z(x, y) =
_
_
_
1
2
H
_
1 cos
_
2
x
L
__
if 0 < x < L and
1
2
B < y <
1
2
B
0 else
(2.2)
where H, B and L denote height, width and length of the obstacle. Potholes are obtained if
negative values for the height (H < 0) are used.
In a similar way track grooves can be modeled too, [48]. By appropriate coordinate transforma-
tions the obstacles can then be integrated into the global road description.
2.2.2 Sine Waves
Using the parallel track road model, a periodic excitation can be realized by
z
1
(s) = A sin (s) , z
2
(s) = A sin (s ) , (2.3)
where s is the path variable, A denotes the amplitude, the wave number, and the angle
describes a phase lag between the left and the right track. The special cases = 0 and =
represent the in-phase excitation with z
1
= z
2
and the out of phase excitation with z
1
= z
2
.
If the vehicle runs with constant velocity ds/dt = v
0
, the momentary position of the vehicle is
given by s = v
0
t, where the initial position s = 0 at t = 0 was assumed. By introducing the
wavelength
L =
2
(2.4)
the term s can be written as
s =
2
L
s =
2
L
v
0
t = 2
v
0
L
t = t . (2.5)
Hence, in the time domain the excitation frequency is given by f = /(2) = v
0
/L.
For most of the vehicles the rigid body vibrations are in between 0.5 Hz to 15 Hz. This range
is covered by waves which satisfy the conditions v
0
/L 0.5 Hz and v
0
/L 15 Hz.
For a given wavelength, lets say L = 4 m, the rigid body vibration of a vehicle are excited if
the velocity of the vehicle will be varied from v
min
0
= 0.5 Hz 4 m = 2 m/s = 7.2 km/h to
v
max
0
= 15 Hz 4 m = 60 m/s = 216 km/h. Hence, to achieve an excitation in the whole
frequency range with moderate vehicle velocities proles with different varying wavelengths
are needed.
2.3 Random Proles
2.3.1 Statistical Properties
Road proles t the category of stationary Gaussian random processes, [6]. Hence, the irreg-
ularities of a road can be described either by the prole itself z
R
= z
R
(s) or by its statistical
properties, Fig. 2.4.
12
2.3 Random Proles
Histogram
Realization
0.15
0.10
0.05
0
-0.05
-0.10
-0.15
-200 -150 -100 -50 0 50 100 150 200
Gaussian
density
function
m
+
[m]
[m]
z
R
s
Figure 2.4: Road prole and statistical properties
By choosing an appropriate reference frame, a vanishing mean value
m = E z
R
(s) = lim
X
1
X
X/2
_
X/2
z
R
(s) ds = 0 (2.6)
can be achieved, where E denotes the expectation operator. Then, the Gaussian density func-
tion which corresponds with the histogram is given by
p(z
R
) =
1
2
e
z
2
R
2
2
, (2.7)
where the deviation or the effective value is obtained from the variance of the process z
R
=
z
R
(s)
2
= E
_
z
2
R
(s)
_
= lim
X
1
X
X/2
_
X/2
z
R
(s)
2
ds . (2.8)
Alteration of effects the shape of the density function. In particular, the points of inexion
occur at . The probability of a value [z[ < is given by
P() =
1
2
+
_
z
2
2
2
dz . (2.9)
In particular, one gets the values: P() = 0.683, P(2) = 0.955, and P(3) = 0.997.
Hence, the probability of a value [z[ 3 is 0.3%.
In extension to the variance of a random process the auto-correlation function is dened by
R() = E z
R
(s) z
R
(s+) = lim
X
1
X
X/2
_
X/2
z
R
(s) z
R
(s+) ds . (2.10)
13
2 Road
The auto-correlation function is symmetric, R() = R(), and it plays an important part in
the stochastic analysis. In any normal random process, as increases the link between z
R
(s)
and z
R
(s+) diminishes. For large values of the two values are practically unrelated. Hence,
R( ) will tend to 0. In fact, R() is always less R(0), which coincides with the variance
2
of the process. If a periodic term is present in the process it will show up in R().
Usually, road proles are characterized in the frequency domain. Here, the auto-correlation
function R() is replaced by the power spectral density (psd) S(). In general, R() and S()
are related to each other by the Fourier transformation
S() =
1
2
R() e
i
d and R() =
S() e
i
d , (2.11)
where i is the imaginary unit, and in rad/m denotes the wave number. To avoid negative
wave numbers, usually a one-sided psd is dened. With
() = 2 S() , if 0 and () = 0 , if < 0 , (2.12)
the relationship e
i
= cos() i sin(), and the symmetry property R() = R()
Eq. (2.11) results in
() =
2
_
0
R() cos () d and R() =
_
0
() cos () d . (2.13)
Now, the variance is obtained from
2
= R( =0) =
_
0
() d . (2.14)
In reality the psd () will be given in a nite interval
1

N
, Fig. 2.5. Then, Eq. (2.14)
1
N
i
(
i
)
Figure 2.5: Power spectral density in a nite interval

can be approximated by a sum, which for N equal intervals will result in
i=1
(
i
) with =

N

1
N
. (2.15)
14
2.3 Random Proles
2.3.2 Classication of Random Road Proles
Road elevation proles can be measured point by point or by high-speed prolometers. The
power spectral densities of roads show a characteristic drop in magnitude with the wave number,
Fig. 2.6a. This simply reects the fact that the irregularities of the road may amount to several
meters over the length of hundreds of meters, whereas those measured over the length of one
meter are normally only some centimeter in amplitude.
Random road proles can be approximated by a psd in the form of
() = (
0
)
_

0
_
w
, (2.16)
where, = 2/L in rad/m denotes the wave number and
0
= (
0
) in m
2
/(rad/m)
describes the value of the psd at a the reference wave number
0
= 1 rad/m. The drop in
magnitude is modeled by the waviness w.
10
-2
10
-1
10
2
10
1
10
0
Wave number [rad/m]
10
-2
10
-1
10
2
10
1
10
0
10
-4
10
-3
10
-5
10
-6
10
-7
10
-8
10
-9 P
o
w
e
r

s
p
e
c
t
r
a
l

d
e
n
s
i
t
y

[
m
2
/
(
r
a
d
/
m
)
]
Wave number [rad/m]
a) Measurements (country road) b) Range of road classes (ISO 8608)
Class A
Class E
0
=25610
6
0
=110
6
Figure 2.6: Road power spectral densities: a) Measurements [3], b) Classication
According to the international directive ISO 8608, [13] typical road proles can be grouped
into classes from A to E. By setting the waviness to w = 2 each class is simply dened by
its reference value
0
. Class A with
0
= 1 10
6
m
2
/(rad/m) characterizes very smooth
highways, whereas Class E with
0
= 256 10
6
m
2
/(rad/m) represents rather rough roads,
Fig. 2.6b.
15
2 Road
2.3.3 Realizations
2.3.3.1 Sinusoidal Approximation
A random prole of a single track can be approximated by a superposition of N sine
waves
z
R
(s) =
N
i=1
A
i
sin (
i
s
i
) , (2.17)
where each sine wave is determined by its amplitude A
i
and its wave number
i
. By different
sets of uniformly distributed phase angles
i
, i = 1(1)N in the range between 0 and 2 different
proles can be generated which are similar in the general appearance but different in details.
The variance of the sinusoidal representation is then given by
2
= lim
X
1
X
X/2
_
X/2
_
N
i=1
A
i
sin (
i
s
i
)
__
N
j=1
A
j
sin (
j
s
j
)
_
ds . (2.18)
For i = j and for i ,= j different types of integrals are obtained. The ones for i = j can be
solved immediately
J
ii
=
_
A
2
i
sin
2
(
i
s
i
) ds =
A
2
i
2
i
_
i
s
i
1
2
sin
_
2 (
i
s
i
)
_
_
. (2.19)
Using the trigonometric relationship
sin x sin y =
1
2
cos(xy)
1
2
cos(x+y) (2.20)
the integrals for i ,= j can be solved too
J
ij
=
_
A
i
sin (
i
s
i
) A
j
sin (
j
s
j
) ds
=
1
2
A
i
A
j
_
cos (
ij
s
ij
) ds
1
2
A
i
A
j
_
cos (
i+j
s
i+j
) ds
=
1
2
A
i
A
j
ij
sin (
ij
s
ij
) +
1
2
A
i
A
j
i+j
sin (
i+j
s
i+j
)
(2.21)
where the abbreviations
ij
=
i
j
and
ij
=
i
j
were used. The sine and cosine
terms in Eqs. (2.19) and (2.21) are limited to values of 1. Hence, Eq. (2.18) simply results in
2
= lim
X
1
X
N
i=1
_
J
ii
X/2
X/2
. .
N
i=1
A
2
i
2
i
i
+ lim
X
1
X
N
i,j=1
_
J
ij
X/2
X/2
. .
0
=
1
2
N
i=1
A
2
i
. (2.22)
16
2.3 Random Proles
On the other hand, the variance of a sinusoidal approximation to a random road prole is given
by Eq. (2.15). So, a road prole z
R
= z
R
(s) described by Eq. (2.17) will have a given psd ()
if the amplitudes are generated according to
A
i
=
_
2 (
i
) , i = 1(1)N , (2.23)
and the wave numbers
i
are chosen to lie at N equal intervals .
0.10
0.05
-0.10
-0.05
0
0 10 20 30 40 50 60 70 80 90 100
[m]
[m]
Road profile z=z(s)
Figure 2.7: Realization of a country road
A realization of the country road with a psd of
0
= 10 10
6
m
2
/(rad/m) is shown in
Fig. 2.7. According to Eq. (2.17) the prole z = z(s) was generated by N = 200 sine waves
in the frequency range from
1
= 0.0628 rad/m to
N
= 62.83 rad/m. The amplitudes A
i
,
i = 1(1)N were calculated by Eq. (2.23) and the MATLAB
function rand was used to

produce uniformly distributed random phase angles in the range between 0 and 2.
2.3.3.2 Shaping Filter
The white noise process produced by random number generators has a uniform spectral density,
and is therefore not suitable to describe real road proles. But, if the white noise process is used
as input to a shaping lter more appropriate spectral densities will be obtained, [29]. A simple
rst order shaping lter for the road prole z
R
reads as
d
ds
z
R
(s) = z
R
(s) +w(s) , (2.24)
where is a constant, and w(s) is a white noise process with the spectral density
w
. Then, the
spectral density of the road prole is obtained from
R
= H()
W
H
T
() =
1
+i

W
1
i
=

W
2
+
2
, (2.25)
where is the wave number, and H() is the frequency response function of the shaping lter.
By setting
W
= 10 10
6
m
2
/(rad/m) and = 0.01 rad/m the measured psd of a typical
country road can be approximated very well, Fig. 2.8.
The shape lter approach is also suitable for modeling parallel tracks, [34]. Here, the cross-
correlation between the irregularities of the left and right track have to be taken into account
too.
17
2 Road
10
-2
10
-1
10
2
10
1
10
0
10
-4
10
-3
10
-5
10
-6
10
-7
10
-8
10
-9 P
o
w
e
r

s
p
e
c
t
r
a
l

d
e
n
s
i
t
y

[
m
2
/
(
r
a
d
/
m
)
]
Wave number [rad/m]
Measurements
Shaping filter
Figure 2.8: Shaping lter as approximation to measured psd
2.3.3.3 Two-Dimensional Model
The generation of fully two-dimensional road proles z
R
= z
R
(x, y) via a sinusoidal approxi-
mation is very laborious. Because a shaping lter is a dynamic system, the resulting road prole
realizations are not reproducible. By adding band-limited white noise processes and taking the
momentary position x, y as seed for the random number generator a reproducible road prole
can be generated, [36].
-4
-2
0
2
4
0
5
10
15
20
25
30
35
40
45
50
-1
0
1
m
z
x
y
Figure 2.9: Two-dimensional road prole
By assuming the same statistical properties in longitudinal and lateral direction two-dimensional
proles, like the one in Fig. 2.9, can be obtained.
18
3 Tire
3.1 Introduction
3.1.1 Tire Development
Some important mile stones in the development of pneumatic tires are shown in Table 3.1.
1839 Charles Goodyear: vulcanization
1845 Robert William Thompson: rst pneumatic tire
(several thin inated tubes inside a leather cover)
1888 John Boyd Dunlop: patent for bicycle (pneumatic) tires
1893 The Dunlop Pneumatic and Tyre Co. GmbH, Hanau, Germany
1895 Andr and Edouard Michelin: pneumatic tires for Peugeot
Paris-Bordeaux-Paris (720 Miles):
50 tire deations,
22 complete inner tube changes
1899 Continental: long-lived tires (approx. 500 Kilometer)
1904 Carbon added: black tires.
1908 Frank Seiberling: grooved tires with improved road traction
1922 Dunlop: steel cord thread in the tire bead
1943 Continental: patent for tubeless tires
1946 Radial Tire
Table 3.1: Milestones in tire development
Of course the tire development did not stop in 1946, but modern tires are still based on this
achievements.
3.1.2 Tire Composites
Tires are very complex. They combine dozens of components that must be formed, assembled
and cured together. And their ultimate success depends on their ability to blend all of the sep-
arate components into a cohesive product that satises the drivers needs. A modern tire is a
mixture of steel, fabric, and rubber. The main composites of a passenger car tire with an overall
mass of 8.5 kg are listed in Table 3.2.
19
3 Tire
Reinforcements: steel, rayon, nylon 16%
Rubber: natural/synthetic 38%
Compounds: carbon, silica, chalk, ... 30%
Softener: oil, resin 10%
Vulcanization: sulfur, zinc oxide, ... 4%
Miscellaneous 2%
Table 3.2: Tire composites: 195/65 R 15 ContiEcoContact, data from www.felge.de
3.1.3 Tire Forces and Torques
In any point of contact between the tire and the road surface normal and friction forces are
transmitted. According to the tires prole design the contact patch forms a not necessarily
coherent area, Fig. 3.1.
180 mm
1
4
0

m
m
Figure 3.1: Tire footprint of a passenger car at normal loading condition: Continental 205/55
R16 90 H, 2.5 bar, F
z
= 4700 N
The effect of the contact forces can be fully described by a resulting force vector applied at a
specic point of the contact patch and a torque vector. The vectors are described in a track-xed
reference frame. The z-axis is normal to the track, the x-axis is perpendicular to the z-axis and
perpendicular to the wheel rotation axis e
yR
. Then, the demand for a right-handed reference
frame also xes the y-axis.
The components of the contact force vector are named according to the direction of the axes,
Fig. 3.2.
Anon symmetric distribution of the forces in the contact patch causes torques around the x and y
axes. A cambered tire generates a tilting torque T
x
. The torque T
y
includes the rolling resistance
of the tire. In particular, the torque around the z-axis is important in vehicle dynamics. It consists
of two parts,
T
z
= T
B
+T
S
. (3.1)
20
3.1 Introduction
F
x
longitudinal force
F
y
lateral force
F
z
vertical force or wheel load
T
x
tilting torque
T
y
rolling resistance torque
T
z
self aligning and bore torque
F
x
F
y
F
z
T
x
T
y
T
z
e
yR
Figure 3.2: Contact forces and torques
The rotation of the tire around the z-axis causes the bore torque T
B
. The self aligning torque
T
S
takes into account that ,in general, the resulting lateral force is not acting in the center of the
contact patch.
3.1.4 Measuring Tire Forces and Torques
To measure tire forces and torques on the road a special test trailer is needed, Fig. 3.4. Here, the
tire
test wheel
compensation wheel
real road
exact contact
Test trailer
Figure 3.3: Layout of a tire test trailer
measurements are performed under real operating conditions. Arbitrary surfaces like asphalt or
concrete and different environmental conditions like dry, wet or icy are possible. Measurements
with test trailers are quite cumbersome and in general they are restricted to passenger car tires.
Indoor measurements of tire forces and torques can be performed on drums or on a at bed,
Fig. 3.4.
21
3 Tire
tire
tire
safety walk
coating
rotation
drum
too small
contact area
too large contact area
tire
safety walk coating perfect contact
Figure 3.4: Drum and at bed tire test rig
On drum test rigs the tire is placed either inside or outside of the drum. In both cases the shape
of the contact area between tire and drum is not correct. That is why, one can not rely on the
measured self aligning torque. Due its simple and robust design, wide applications including
measurements of truck tires are possible.
The at bed tire test rig is more sophisticated. Here, the contact patch is as at as on the road.
But, the safety walk coating which is attached to the steel bed does not generate the same friction
conditions as on a real road surface.
-40 -30 -20 -10 0 10 20 30 40
Longitudinal slip [%]
-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
L
o
n
g
i
t
u
d

f
o
r
c
e

F
x

[
N
]
Radial 205/50 R15, F
N
= 3500 N, dry asphalt
Driving
Braking
Figure 3.5: Typical results of tire measurements
22
3.1 Introduction
Tire forces and torques are measured in quasi-static operating conditions. Hence, the measure-
ments for increasing and decreasing the sliding conditions usually result in different graphs,
Fig. 3.5. In general, the mean values are taken as steady state results.
3.1.5 Modeling Aspects
For the dynamic simulation of on-road vehicles, the model-element tire/road is of special im-
portance, according to its inuence on the achievable results. It can be said that the sufcient
description of the interactions between tire and road is one of the most important tasks of vehicle
modeling, because all the other components of the chassis inuence the vehicle dynamic prop-
erties via the tire contact forces and torques. Therefore, in the interest of balanced modeling, the
precision of the complete vehicle model should stand in reasonable relation to the performance
of the applied tire model. At present, two groups of models can be identied, handling models
and structural or high frequency models, [18].
Structural tire models are very complex. Within RMOD-K [25] the tire is modeled by four
circular rings with mass points that are also coupled in lateral direction. Multi-track contact and
the pressure distribution across the belt width are taken into account. The tire model FTire [9]
consists of an extensible and exible ring which is mounted to the rim by distributed stiffnesses
in radial, tangential and lateral direction. The ring is approximated by a nite number of belt
elements to which a number of mass-less tread blocks are assigned, Fig. 3.6.
c
long.
c
bend. in-plane
c
bend. out-of- plane
c
torsion
F
Frict.
c
Frict.
c
dyn.
d
dyn.
d
rad.
c
rad.
belt node
rim
Model
Structure
Radial
Force
Element
(v,p,T)
x, v
x
B
, v
B
Contact
Element
Figure 3.6: Complex tire model (FTire)
Complex tire models are computer time consuming and they need a lot a data. Usually, they are
used for stochastic vehicle vibrations occurring during rough road rides and causing strength-
relevant component loads, [32].
Comparatively lean tire models are suitable for vehicle dynamics simulations, while, with the
exception of some elastic partial structures such as twist-beam axles in cars or the vehicle frame
23
3 Tire
in trucks, the elements of the vehicle structure can be seen as rigid. On the tires side, semi-
physical tire models prevail, where the description of forces and torques relies, in contrast
to purely physical tire models, also on measured and observed force-slip characteristics. This
class of tire models is characterized by an useful compromise between user-friendliness, model-
complexity and efciency in computation time on the one hand, and precision in representation
on the other hand.
In vehicle dynamic practice often there exists the problem of data provision for a special type of
tire for the examined vehicle. Considerable amounts of experimental data for car tires has been
published or can be obtained from the tire manufacturers. If one cannot nd data for a special
tire, its characteristics can be guessed at least by an engineers interpolation of similar tire types,
Fig. 3.7. In the eld of truck tires there is still a considerable backlog in data provision. These
circumstances must be respected in conceiving a user-friendly tire model.
F
y
sx
s
s
y
S
F
S
M
F
M
dF
0
F(s)
dF
S
y
F
y
F
y
M
S
s
y
M
s
y
0
F
y
s
y
dFx
0
F
x
M
F
x
S
F
x
s
x
M
s
x
S
s
x
F
x
s
s
d
y
c
y
F
y
v
y
Q P
y
e
Dynamic Forces
Combined Forces
Figure 3.7: Handling tire model: TMeasy [11]
For a special type of tire, usually the following sets of experimental data are provided:
longitudinal force versus longitudinal slip (mostly just brake-force),
lateral force versus slip angle,
aligning torque versus slip angle,
radial and axial compliance characteristics,
whereas additional measurement data under camber and low road adhesion are favorable special
cases.
Any other correlations, especially the combined forces and torques, effective under operating
conditions, often have to be generated by appropriate assumptions with the model itself, due to
the lack of appropriate measurements. Another problem is the evaluation of measurement data
from different sources (i.e. measuring techniques) for a special tire, [12]. It is a known fact that
24
3.2 Contact Geometry
different measuring techniques result in widely spread results. Here the experience of the user
is needed to assemble a probably best set of data as a basis for the tire model from these sets
of data, and to verify it eventually with own experimental results.
3.2.1 Basic Approach
The current position of a wheel in relation to the xed x
0
-, y
0
- z
0
-system is given by the wheel
center M and the unit vector e
yR
in the direction of the wheel rotation axis, Fig. 3.8.
road: z = z ( x , y )
e
yR
M
e
n
0
P
tire
x
0
0
y
0
z
0
*
P
P
x
0
0
y
0
z
0
e
yR
M
e
n
e
x
e
y
rim
centre
plane
local road plane
e
zR
r
MP
wheel
carrier
0
P
a
b
Figure 3.8: Contact geometry
The irregularities of the track can be described by an arbitrary function of two spatial coordi-
nates
z = z(x, y). (3.2)
At an uneven track the contact point P can not be calculated directly. At rst, one can get an
estimated value with the vector r
MP
= r
0
e
zB
, where r
0
is the undeformed tire radius, and
e
zB
is the unit vector in the z-direction of the body xed reference frame. Usually, the point P
does not lie on the track. The corresponding track point P

0
can be calculated via Eq. (3.2). In the
point P
0
the track normal e
n
is calculated, now. Then the unit vectors in the tires circumferential
direction and lateral direction can be determined.
The tire camber angle
= arcsin
_
e
T
yR
e
n
_
(3.3)
25
3 Tire
describes the inclination of the wheel rotation axis against the track normal.
The vector from the rim center M to the track point P
0
is split into three parts now
r
MP
0
= r
S
e
zR
+a e
x
+b e
y
, (3.4)
where r
S
denotes the loaded or static tire radius, a, b are distances measured in circumferential
and lateral direction, and the radial direction is given by the unit vector
e
zR
= e
x
e
yR
(3.5)
which is perpendicular to e
x
and e
yR
. A scalar multiplication of Eq. (3.4) with e
n
results in
e
T
n
r
MP
0
= r
S
e
T
n
e
zR
+a e
T
n
e
x
+b e
T
n
e
y
. (3.6)
As the unit vectors e
x
and e
y
are perpendicular to e
n
Eq. (3.6) simplies to
e
T
n
r
MP
0
= r
S
e
T
n
e
zR
. (3.7)
Hence, the static tire radius is given by
r
S
=
e
T
n
r
MP
0
e
T
n
e
zR
. (3.8)
The contact point P given by the vector
r
MP
= r
S
e
zR
(3.9)
lies within the rim center plane. The transition from the point P
0
to the contact point P takes
place according to Eq. (3.4) by the terms a e
x
and b e
y
perpendicular to the track normal e
n
. The
track normal, however, was calculated in the point P
0
. With an uneven track the point P no
longer lies on the track and can therefor no longer considered exactly as contact point.
With the newly estimated value P
= P the calculations may be repeated until the difference

between P and P
0
is sufciently small. Tire models which can be simulated within acceptable
time assume that the contact patch is sufciently at. At an ordinary passenger car tire, the
contact area has at normal load approximately the size of 1520 cm. Hence, it makes no sense
to calculate a ctitious contact point to fractions of millimeters, when later on the real track
will be approximated by a plane in the range of centimeters. If the track in the contact area is
replaced by a local plane, no further iterative improvements will be necessary for the contact
point calculation.
3.2.2 Tire Deection
For a vanishing camber angle = 0 the deected zone has a rectangular shape, Fig. 3.9. Its area
is given by
A
0
= z b , (3.10)
26
r
S
r
0
e
yR
e
n
P
z
w
C
= b
r
SL
r
0
e
yR
e
n
P
b
r
SR
r
0
e
yR
e
n
P
b*
r
SR
full contact partial contact

= 0
= 0
w
C
w
C
/
r
S
r
S
Figure 3.9: Tire deection
where b is the width of the tire, and the tire deection is obtained by
z = r
0
r
S
. (3.11)
Here, the width of the tire simply equals the width of the contact zone, w
C
= b.
On a cambered tire the deected zone of the tire cross section depends on the contact situation.
The magnitude of the tire ank radii
r
SL
= r
s
+
b
2
tan and r
SR
= r
s
b
2
tan (3.12)
determines the shape of the deected zone.
The tire will be in full contact to the road if r
SL
r
0
and r
SR
r
0
hold. Then, the deected
zone has a trapezoidal shape with an area of
A
=
1
2
(r
0
r
SR
+r
0
r
SL
) b = (r
0
r
S
) b . (3.13)
Equalizing the cross sections A
0
= A
results in
z = r
0
r
S
. (3.14)
Hence, at full contact the tire camber angle has no inuence on the vertical tire force. But,
due to
w
C
=
b
cos
(3.15)
the width of the contact area increases with the tire camber angle.
27
3 Tire
The deected zone will change to a triangular shape if one of the ank radii exceeds the unde-
ected tire radius. Assuming r
SL
> r
0
and r
SR
< r
0
the area of the deected zone is obtained
by
A
=
1
2
(r
0
r
SR
) b
, (3.16)
where the width of the deected zone follows from
b
=
r
0
r
SR
tan
. (3.17)
Now, Eq. (3.16) reads as
A
=
1
2
(r
0
r
SR
)
2
tan
. (3.18)
Equalizing the cross sections A
0
= A
results in
z =
1
2
_
r
0
r
S
+
b
2
tan
_
2
b tan
. (3.19)
where Eq. (3.12) was used to express the ank radius r
SR
by the static tire radius r
S
, the tire
width b and the camber angle . Now, the width of the contact area is given by
w
C
=
b
cos
=
r
0
r
SR
tan cos
=
r
0
r
S
+
b
2
tan
sin
, (3.20)
where the Eqs. (3.17) and (3.12) where used to simplify the expression. If tan and sin
are replaced by [ tan [ and [ sin [ then, the Eqs. (3.19) and (3.20) will hold for positive and
negative camber angles.
3.2.3 Length of Contact Patch
To approximate the length of the contact patch the tire deformation is split into two parts,
Fig. 3.10. By z
F
and z
B
the average tire ank and the belt deformation are measured. Hence,
for a tire with full contact to the road
z = z
F
+z
B
= r
0
r
S
(3.21)
will hold.
Assuming both deections being equal will lead to
z
F
z
B

1
2
z . (3.22)
Approximating the belt deection by truncating a circle with the radius of the undeformed tire
results in
_
L
2
_
2
+ (r
0
z
B
)
2
= r
2
0
. (3.23)
28
F
z
L
r
0
r
S
Belt
Rim
L/2
r
0
z
F
z
B
z
B
undeformed
belt
Figure 3.10: Length of contact patch
In normal driving situations the belt deections are small, z
B
r
0
. Hence, Eq. (3.23) can be
simplied and nally results in
L
2
4
= 2 r
0
z
B
or L =
_
8 r
0
z
B
. (3.24)
Inspecting the passenger car tire footprint in Fig. 3.1 leads to a contact patch length of
L 140 mm. For this tire the radial stiffness and the inated radius are specied with c
R
=
265 000N/mand r
0
= 316.9mm. The overall tire deection can be estimated by z = F
z
/c
R
.
At the load of F
z
= 4700N the deection amounts to z = 4700N / 265 000N/m = 0.0177m.
Then, by approximating the belt deformation by the half of the tire deection, the length of the
contact patch will become L =
_
8 0.3169 m 0.0177/2 m = 0.1498 m 150 mm which
corresponds quite well with the length of the tire footprint.
3.2.4 Static Contact Point
Assuming that the pressure distribution on a cambered tire with full road contact corresponds
with the trapezoidal shape of the deected tire area, the acting point of the resulting vertical
tire force F
Z
will be shifted from the geometric contact point P to the static contact point
Q, Fig. 3.11. If the cambered tire has only a partial contact to the road then, according to the
deection area a triangular pressure distribution will be assumed.
The center of the trapezoidal area or, in the case of a partial contact the center of the triangle,
determines the lateral deviation y
Q
. The static contact point Q described by the vector
r
0Q
= r
0P
+ y
Q
e
y
(3.25)
represents the contact patch much better than the geometric contact point P.
29
3 Tire
e
n
P
w
C
r
S
Q
F
z
r
0
-r
SL
r
0
-r
SR
y
e
y
A
A
e
n
P
w
C
Q
F
z
y
e
y
b/2
Figure 3.11: Lateral deviation of contact point at full and partial contact
3.2.5 Contact Point Velocity
To calculate the tire forces and torques which are generated by friction the contact point velocity
will be needed. The static contact point Q given by Eq. (3.25) can be expressed as follows
r
0Q
= r
0M
+ r
MQ
, (3.26)
where M denotes the wheel center and hence, the vector r
MQ
describes the position of static
contact point Q relative to the wheel center M. The absolute velocity of the contact point will
be obtained from
v
0Q,0
= r
0Q,0
= r
0M,0
+ r
MQ,0
, (3.27)
where r
0M,0
= v
0M,0
denotes the absolute velocity of the wheel center. The vector r
MQ
takes
part on all those motions of the wheel carrier which do not contain elements of the wheel
rotation and it In addition, it contains the tire deection z normal to the road. Hence, its time
derivative can be calculated from
r
MQ,0
=
0R,0
r
MQ,0
+ z e
n,0
, (3.28)
where
0R
is the angular velocity of the wheel rim without any component in the direction of
the wheel rotation axis, z denotes the change of the tire deection, and e
n
describes the road
normal. Now, Eq. (3.27) reads as
v
0Q,0
= v
0M,0
+
0R,0
r
MQ,0
+ z e
n,0
. (3.29)
As the point Q lies on the track, v
0Q,0
must not contain any component normal to the track
e
T
n,0
v
0P,0
= 0 or e
T
n,0
_
v
0M,0
+
0R,0
r
MQ,0
_
+ z e
T
n,0
e
n,0
= 0 . (3.30)
As e
n,0
is a unit vector, e
T
n,0
e
n,0
= 1 will hold, and then, the time derivative of the tire deforma-
tion is simply given by
z = e
T
n,0
_
v
0M,0
+
0R,0
r
MQ,0
_
. (3.31)
30
Finally, the components of the contact point velocity in longitudinal and lateral direction are
obtained from
v
x
= e
T
x,0
v
0Q,0
= e
T
x,0
_
v
0M,0
+
0R,0
r
MQ,0
_
(3.32)
and
v
y
= e
T
y,0
v
0P,0
= e
T
y,0
_
v
0M,0
+
0R,0
r
MQ,0
_
, (3.33)
where the relationships e
T
x,0
e
n,0
= 0 and e
T
y,0
e
n,0
= 0 were used to simplify the expressions.
3.2.6 Dynamic Rolling Radius
At an angular rotation of , assuming the tread particles stick to the track, the deected tire
moves on a distance of x, Fig. 3.12.
x
r
0
r
S

r
x

D
deflected tire rigid wheel

v
t
Figure 3.12: Dynamic rolling radius
With r
0
as unloaded and r
S
= r
0
r as loaded or static tire radius
r
0
sin = x (3.34)
and
r
0
cos = r
S
(3.35)
hold.
If the motion of a tire is compared to the rolling of a rigid wheel, then, its radius r
D
will have
to be chosen so that at an angular rotation of the tire moves the distance
r
0
sin = x = r
D
. (3.36)
Hence, the dynamic tire radius is given by
r
D
=
r
0
sin
. (3.37)
For 0 one obtains the trivial solution r
D
= r
0
.
31
3 Tire
At small, yet nite angular rotations the sine-function can be approximated by the rst terms of
its Taylor-Expansion. Then, Eq. (3.37) reads as
r
D
= r
0

1
6
= r
0
_
1
1
6
2
_
. (3.38)
With the according approximation for the cosine-function
r
S
r
0
= cos = 1
1
2
2
or
2
= 2
_
1
r
S
r
0
_
(3.39)
one nally gets
r
D
= r
0
_
1
1
3
_
1
r
S
r
0
__
=
2
3
r
0
+
1
3
r
S
. (3.40)
Due to r
S
= r
S
(F
z
) the ctive radius r
D
depends on the wheel load F
z
. Therefore, it is called
dynamic tire radius. If the tire rotates with the angular velocity , then
v
t
= r
D
(3.41)
will denote the average velocity at which the tread particles are transported through the contact
patch.
3.3 Forces and Torques caused by Pressure
Distribution
3.3.1 Wheel Load
The vertical tire force F
z
can be calculated as a function of the normal tire deection z and
the deection velocity z
F
z
= F
z
(z, z) . (3.42)
Because the tire can only apply pressure forces to the road the normal force is restricted to
F
z
0. In a rst approximation F
z
is separated into a static and a dynamic part
F
z
= F
S
z
+F
D
z
. (3.43)
The static part is described as a nonlinear function of the normal tire deection
F
S
z
= a
1
z + a
2
(z)
2
. (3.44)
The constants a
1
and a
2
may be calculated from the radial stiffness at nominal and double
payload.
Results for a passenger car and a truck tire are shown in Fig. 3.13. The parabolic approximation
in Eq. (3.44) ts very well to the measurements. The radial tire stiffness of the passenger car
32
3.3 Forces and Torques caused by Pressure Distribution
0 10 20 30 40 50
0
2
4
6
8
10
Passenger Car Tire: 205/50 R15
F
z

[
k
N
]
0 20 40 60 80
0
20
40
60
80
100
Truck Tire: X31580 R22.5
F
z

[
k
N
]
z [mm]
z [mm]
Figure 3.13: Tire radial stiffness: Measurements, Approximation
tire at the payload of F
z
= 3 200 N can be specied with c
0
= 190 000N/m. The Payload
F
z
= 35 000 N and the stiffness c
0
= 1 250 000N/m of a truck tire are signicantly larger.
The dynamic part is roughly approximated by
F
D
z
= d
R
z , (3.45)
where d
R
is a constant describing the radial tire damping, and the derivative of the tire defor-
mation z is given by Eq. (3.31).
3.3.2 Tipping Torque
The lateral shift of the vertical tire force F
z
from the geometric contact point P to the static
contact point Q is equivalent to a force applied in P and the tipping torque
M
x
= F
z
y
Q
(3.46)
acting around a longitudinal axis in P, Fig. 3.14.
e
n
P Q
F
z
y
e
y
e
n
P
F
z
e
y
T
x
Figure 3.14: Tipping torque at full contact

Note: Fig. 3.14 shows a negative tipping torque. Because a positive camber angle moves the
contact point into the negative y-direction and hence, will generate a negative tipping torque.
33
3 Tire
e
n
P
Q
F
z
y
e
y
Figure 3.15: Cambered tire with partial contact
The use of the tipping torque instead of shifting the contact point is limited to those cases where
the tire has full or nearly full contact to the road. If the cambered tire has only partly contact to
the road, the geometric contact point P may even be located outside the contact area whereas
the static contact point Q is still a real contact point, Fig. 3.15.
3.3.3 Rolling Resistance
If a non-rotating tire has contact to a at ground the pressure distribution in the contact patch
will be symmetric from the front to the rear, Fig. 3.16. The resulting vertical force F
z
is applied
in the center C of the contact patch and hence, will not generate a torque around the y-axis.
F
z
C
F
z
C
e
x
e
n
rotating
e
x
e
n
non-rotating
x
R
Figure 3.16: Pressure distribution at a non-rotation and rotation tire
If the tire rotates tread particles will be stuffed into the front of the contact area which causes
a slight pressure increase, Fig. 3.16. Now, the resulting vertical force is applied in front of the
contact point and generates the rolling resistance torque
T
y
= F
z
x
R
sign() , (3.47)
where sign() assures that T
y
always acts against the wheel angular velocity . The distance
x
R
from C to the working point of F
z
usually is related to the unloaded tire radius r
0
f
R
=
x
R
r
0
. (3.48)
According to [20] the dimensionless rolling resistance coefcient slightly increases with the
traveling velocity v of the vehicle
f
R
= f
R
(v) . (3.49)
34
3.4 Friction Forces and Torques
Under normal operating conditions, 20km/h < v < 200km/h, the rolling resistance coefcient
for typical passenger car tires is in the range of 0.01 < f
R
< 0.02.
The rolling resistance hardly inuences the handling properties of a vehicle. But it plays a major
part in fuel consumption.
3.4.1 Longitudinal Force and Longitudinal Slip
To get a certain insight into the mechanism generating tire forces in longitudinal direction, we
consider a tire on a at bed test rig. The rim rotates with the angular velocity and the at bed
runs with the velocity v
x
. The distance between the rim center and the at bed is controlled to
the loaded tire radius corresponding to the wheel load F
z
, Fig. 3.17.
A tread particle enters at the time t = 0 the contact patch. If we assume adhesion between
the particle and the track, then the top of the particle will run with the bed velocity v
x
and the
bottom with the average transport velocity v
t
= r
D
. Depending on the velocity difference
v = r
D
v
x
the tread particle is deected in longitudinal direction
u = (r
D
v
x
) t . (3.50)
v
x
L
r
D
u
u
max
r
D
v
x
Figure 3.17: Tire on at bed test rig
The time a particle spends in the contact patch can be calculated by
T =
L
r
D
[[
, (3.51)
where L denotes the contact length, and T > 0 is assured by [[.
35
3 Tire
The maximum deection occurs when the tread particle leaves the contact patch at the time
t = T
u
max
= (r
D
v
x
) T = (r
D
v
x
)
L
r
D
[[
. (3.52)
The deected tread particle applies a force to the tire. In a rst approximation we get
F
t
x
= c
t
x
u , (3.53)
where c
t
x
represents the stiffness of one tread particle in longitudinal direction.
On normal wheel loads more than one tread particle is in contact with the track, Fig. 3.18a. The
number p of the tread particles can be estimated by
p =
L
s +a
, (3.54)
where s is the length of one particle and a denotes the distance between the particles.
c u
b)
L
max
t
x
*
c u
t
u
*
a)
L
s
a
Figure 3.18: a) Particles, b) Force distribution,
Particles entering the contact patch are undeformed, whereas the ones leaving have the max-
imum deection. According to Eq. (3.53), this results in a linear force distribution versus the
contact length, Fig. 3.18b. The resulting force in longitudinal direction for p particles is given
by
F
x
=
1
2
p c
t
x
u
max
. (3.55)
Using the Eqs. (3.54) and (3.52) this results in
F
x
=
1
2
L
s +a
c
t
x
(r
D
v
x
)
L
r
D
[[
. (3.56)
A rst approximation of the contact length L was calculated in Eq. (3.24). Approximating the
belt deformation by z
B

1
2
F
z
/c
R
results in
L
2
4 r
0
F
z
c
R
, (3.57)
where c
R
denotes the radial tire stiffness, and nonlinearities and dynamic parts in the tire defor-
mation were neglected. Now, Eq. (3.55) can be written as
F
x
= 2
r
0
s +a
c
t
x
c
R
F
z
r
D
v
x
r
D
[[
. (3.58)
36
The nondimensional relation between the sliding velocity of the tread particles in longitudinal
direction v
S
x
= v
x
r
D
and the average transport velocity r
D
[[ form the longitudinal slip
s
x
=
(v
x
r
D
)
r
D
[[
. (3.59)
The longitudinal force F
x
is proportional to the wheel load F
z
and the longitudinal slip s
x
in
this rst approximation
F
x
= k F
z
s
x
, (3.60)
where the constant k summarizes the tire properties r
0
, s, a, c
t
x
and c
R
.
Equation (3.60) holds only as long as all particles stick to the track. At moderate slip values the
particles at the end of the contact patch start sliding, and at high slip values only the parts at the
beginning of the contact patch still stick to the road, Fig. 3.19.
L
adhesion
F
x
t
<
=
F
H
t
small slip values
F = k F s
x
* * x F = F f ( s )
x * x F = F
x G
z z
L
adhesion
F
x
t
F
H
t
moderate slip values
L
sliding
F
x
t
F
G
high slip values
=
sliding
=
Figure 3.19: Longitudinal force distribution for different slip values
The resulting nonlinear function of the longitudinal force F
x
versus the longitudinal slip s
x
can be dened by the parameters initial inclination (driving stiffness) dF
0
x
, location s
M
x
and
magnitude of the maximum F
M
x
, start of full sliding s
S
x
and the sliding force F
S
x
, Fig. 3.20.
F
x
x
M
x
S
dF
x
0
s
x
s
x
s
x
M S
F
F
adhesion
sliding
Figure 3.20: Typical longitudinal force characteristics
37
3 Tire
3.4.2 Lateral Slip, Lateral Force and Self Aligning Torque
Similar to the longitudinal slip s
x
, given by Eq. (3.59), the lateral slip can be dened by
s
y
=
v
S
y
r
D
[[
, (3.61)
where the sliding velocity in lateral direction is given by
v
S
y
= v
y
(3.62)
and the lateral component of the contact point velocity v
y
follows from Eq. (3.33).
As long as the tread particles stick to the road (small amounts of slip), an almost linear distri-
bution of the forces along the length L of the contact patch appears. At moderate slip values the
particles at the end of the contact patch start sliding, and at high slip values only the parts at the
beginning of the contact patch stick to the road, Fig. 3.21.
L
a
d
h
e
s
i
o
n
F
y
small slip values
L
a
d
h
e
s
i
o
n
F
y
s
l
i
d
i
n
g
moderate slip values
L
s
l
i
d
i
n
g
F
y
large slip values
n
F = k F s
y * * y F = F f ( s )
y
* y F = F
y G
z
z
Figure 3.21: Lateral force distribution over contact patch
The nonlinear characteristics of the lateral force versus the lateral slip can be described by
the initial inclination (cornering stiffness) dF
0
y
, the location s
M
y
and the magnitude F
M
y
of the
maximum, the beginning of full sliding s
S
y
, and the magnitude F
S
y
of the sliding force.
The distribution of the lateral forces over the contact patch length also denes the point of
application of the resulting lateral force. At small slip values this point lies behind the center
of the contact patch (contact point P). With increasing slip values it moves forward, sometimes
even before the center of the contact patch. At extreme slip values, when practically all particles
are sliding, the resulting force is applied at the center of the contact patch.
The resulting lateral force F
y
with the dynamic tire offset or pneumatic trail n as a lever gener-
ates the self aligning torque
T
S
= nF
y
. (3.63)
The lateral force F
y
as well as the dynamic tire offset are functions of the lateral slip s
y
.
Typical plots of these quantities are shown in Fig. 3.22. Characteristic parameters of the lateral
38
F
y
y
M
y
S
dF
y
0
s
y s
y
s
y
M S
F
F
adhesion
adhesion/
sliding
full sliding
adhesion
adhesion/sliding
n/L
0
s
y
s
y
S
s
y
0
(n/L)
adhesion
adhesion/sliding
M
s
y
s
y
S
s
y
0
S
full sliding
full sliding
Figure 3.22: Typical plot of lateral force, tire offset and self aligning torque
force graph are initial inclination (cornering stiffness) dF
0
y
, location s
M
y
and magnitude of the
maximum F
M
y
, begin of full sliding s
S
y
, and the sliding force F
S
y
.
The dynamic tire offset has been normalized by the length of the contact patch L. The initial
value (n/L)
0
as well as the slip values s
0
y
and s
S
y
sufciently characterize the graph.
The normalized dynamic tire offset starts at s
y
= 0 with an initial value (n/L)
0
> 0 and, it
tends to zero, n/L 0 at large slip values, s
y
s
S
y
. Sometimes the normalized dynamic tire
offset overshoots to negative values before it reaches zero again.
The value of (n/L)
0
can be estimated very well. At small values of lateral slip s
y
0 one gets
in a rst approximation a triangular distribution of lateral forces over the contact area length cf.
Fig. 3.21. The working point of the resulting force (dynamic tire offset) is then given by
n(F
z
0, s
y
=0) =
1
6
L . (3.64)
The value n =
1
6
L can only serve as reference point, for the uneven distribution of pressure
in longitudinal direction of the contact area results in a change of the deexion prole and the
dynamic tire offset.
3.4.3 Wheel Load Inuence
The resistance of a real tire against deformations has the effect that with increasing wheel load
the distribution of pressure over the contact area becomes more and more uneven. The tread
particles are deected just as they are transported through the contact area. The pressure peak
in the front of the contact area cannot be used, for these tread particles are far away from the
adhesion limit because of their small deection. In the rear of the contact area the pressure drop
leads to a reduction of the maximally transmittable friction force. With rising imperfection of
the pressure distribution over the contact area, the ability to transmit forces of friction between
tire and road lessens.
39
3 Tire
-0.4 -0.2 0 0.2 0.4
-6
-4
-2
0
2
4
6
-20 -10 0 10 20
-6
-4
-2
0
2
4
6
F
x

[
k
N
]
s
x
[-]
F
y

[
k
N
]
[deg]
Figure 3.23: Longitudinal and lateral force characteristics: F
z
= 1.8, 3.2, 4.6, 5.4, 6.0 kN
In practice, this leads to a digressive inuence of the wheel load on the characteristic curves of
the longitudinal force and in particular of the lateral force, Fig. 3.23.
3.4.4 Different Friction Coefcients
The tire characteristics are valid for one specic tire road combination only.
-0.5 0
0.5
-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
F

s
y
[-]
F
y

[
N
]
L
/
0
0.2
0.4
0.6
0.8
1.0
F
z
= 3.2 kN
Figure 3.24: Lateral force characteristics for different friction coefcients
A reduced or changed friction coefcient mainly inuences the maximum force and the sliding
force, whereas the initial inclination will remain unchanged, Fig. 3.24.
If the road model provides not only the roughness information z = f
R
(x, y) but also the local
friction coefcient [z,
L
] = f
R
(x, y) then, braking on -split maneuvers can easily be simu-
lated, [40].
40
3.4.5 Typical Tire Characteristics
The tire model TMeasy [11] can be used for passenger car tires as well as for truck tires. It
-40 -20 0 20 40
-6
-4
-2
0
2
4
6
F
x

[
k
N
]
1.8 kN
3.2 kN
4.6 kN
5.4 kN
-40 -20 0 20 40
-40
-20
0
20
40
F
x

[
k
N
]
10 kN
20 kN
30 kN
40 kN
50 kN
Passenger car tire
Truck tire
s
x
[%]
s
x
[%]
Figure 3.25: Longitudinal force: Meas., TMeasy
-6
-4
-2
0
2
4
6
F
y

[
k
N
]
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-20 -10 0 10 20
[
o
]
-40
-20
0
20
40
F
y

[
k
N
]
10 kN
20 kN
30 kN
40 kN
-20 -10 0 10 20
[
o
]
Passenger car tire
Truck tire
Figure 3.26: Lateral force: Meas., TMeasy
approximates the characteristic curves F
x
= F
x
(s
x
), F
y
= F
y
() and M
z
= M
z
() quite well
even for different wheel loads F
z
, Figs. 3.25 and ??.
When experimental tire values are missing, the model parameters can be pragmatically esti-
mated by adjustment of the data of similar tire types. Furthermore, due to their physical sig-
nicance, the parameters can subsequently be improved by means of comparisons between the
simulation and vehicle testing results as far as they are available.
41
3 Tire
-20 -10 0 10 20
-150
-100
-50
0
50
100
150
[
o
]
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-20 -10 0 10 20
-1500
-1000
-500
0
500
1000
1500
18.4 kN
36.8 kN
55.2 kN
[
o
]
Passenger car tire
Truck tire
T
z

[
N
m
]
T
z

[
N
m
]
Figure 3.27: Self aligning torque: Meas., TMeasy
3.4.6 Combined Slip
The longitudinal force as a function of the longitudinal slip F
x
= F
x
(s
x
) and the lateral force
depending on the lateral slip F
y
= F
y
(s
y
) can be dened by their characteristic parameters
initial inclination dF
0
x
, dF
0
y
, location s
M
x
, s
M
y
and magnitude of the maximum F
M
x
, F
M
y
as well
as sliding limit s
S
x
, s
S
y
and sliding force F
S
x
, F
S
y
, Fig. 3.28. During general driving situations, e.g.
acceleration or deceleration in curves, longitudinal s
x
and lateral slip s
y
appear simultaneously.
F
y
s
x
s
s
y
S
F
S
M
F
M
dF
0
F(s)
dF
S
y
F
y
F
y
M
S
s
y
M
s
y
0
F
y
s
y
dF
x
0
F
x
M
F
x
S
F
x
s
x
M
s
x
S
s
x
F
x
s
s
Figure 3.28: Generalized tire characteristics
42
The longitudinal slip s
x
and the lateral slip s
y
can vectorially be added to a generalized slip.
Similar to the graphs of the longitudinal and lateral forces the graph F = F(s) of the gen-
eralized tire force can be dened by the characteristic parameters dF
0
, s
M
, F
M
, s
S
and F
S
.
These parameters can be calculated from the corresponding values of the longitudinal and lat-
eral force characteristics. The longitudinal and the lateral forces follow then from the according
projections in longitudinal and lateral direction.
Passenger car tire: F
z
= 3.2 kN Truck tire: F
z
= 35 kN
-4 -2 0 2 4
-3
-2
-1
0
1
2
3
F
x
[kN]
F
y

[
k
N
]
-20 0 20
-30
-20
-10
0
10
20
30
F
x
[kN]
F
y

[
k
N
]
[s
x
[ = 1, 2, 4, 6, 10, 15 %; [[ = 1, 2, 4, 6, 10, 14
Figure 3.29: Two-dimensional tire characteristics

Within the TMeasy model approach one-dimensional characteristics are automatically con-
verted to two-dimensional combined-slip characteristics, Fig. 3.29.
3.4.7 Camber Inuence
At a cambered tire, Fig. 3.30, the angular velocity of the wheel has a component normal to
the road
n
= sin . (3.65)
Now, the tread particles in the contact patch possess a lateral velocity which depends on their
position and is provided by
v
() =
n
L
2
L/2
, = sin , L/2 L/2 . (3.66)
At the contact point it vanishes whereas at the end of the contact patch it takes on the same
value as at the beginning, however, pointing into the opposite direction. Assuming that the tread
particles stick to the track, the deection prole is dened by
y
() = v
() . (3.67)
43
3 Tire
e
yR
v()
rim
centre
plane
y()
r
D
|| e
x
e
y
e
n F = F
y y
(s
y
): Parameter
-0.5 0
0.5
-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
Figure 3.30: Cambered tire F
y
() at F
z
= 3.2 kN and = 0
, 2
, 4
, 6
, 8
The time derivative can be transformed to a space derivative

y
() =
d y
()
d
d
d t
=
d y
()
d
r
D
[[ (3.68)
where r
D
[[ denotes the average transport velocity. Now, Eq. (3.67) can be written as
d y
()
d
r
D
[[ = sin or
d y
()
d
=
sin
r
D
[[
L
2
L/2
, (3.69)
where L/2 was used to achieve dimensionless terms. Similar to the lateral slip s
y
which is
dened by Eq. (3.61) we can introduce a camber slip now
s
=
sin
r
D
[[
L
2
. (3.70)
Then, Eq. (3.69) simplies to
d y
()
d
= s
L/2
. (3.71)
The shape of the lateral displacement prole is obtained by integration
y
= s
1
2
L
2
_

L/2
_
2
+ C . (3.72)
The boundary condition y
_
=
1
2
L
_
= 0 can be used to determine the integration constant C.
One gets
C = s
1
2
L
2
. (3.73)
44
Then, Eq. (3.72) reads as
y
() = s
1
2
L
2
_
1
_

L/2
_
2
_
. (3.74)
The lateral displacements of the tread particles caused by a camber slip are compared now with
the ones caused by pure lateral slip, Fig. 3.31. At a tire with pure lateral slip each tread particle
y()
y
-L/2 0 L/2
y
y
()
y
-L/2 0 L/2
a) camber slip b) lateral slip
y
y
_
y
_
Figure 3.31: Displacement proles of tread particles
in the contact patch possesses the same lateral velocity which results in dy
y
/d r
D
[[ = v
y
,
where according to Eq. (3.68) the time derivative y
y
was transformed to the space derivative
dy
y
/d . Hence, the deection prole is linear, and reads as y
y
= v
y
/(r
D
[[) = s
y
, where
the denition in Eq. (3.61) was used to introduce the lateral slip s
y
. Then, the average deection
of the tread particles under pure lateral slip is given by
y
y
= s
y
L
2
. (3.75)
The average deection of the tread particles under pure camber slip is obtained from
y
= s
1
2
L
2
1
L
L/2
_
L/2
_
1
_
x
L/2
_
2
_
d =
1
3
s
L
2
. (3.76)
A comparison of Eq. (3.75) with Eq. (3.76) shows, that by using
s
y
=
1
3
s
(3.77)
the lateral camber slip s
can be converted to an equivalent lateral slip s
y
.
In normal driving conditions, the camber angle and thus, the lateral camber slip are limited to
small values, s
y
1. So, the lateral camber force can be modeled by
F
y
=
dF
y
s
y
s
y
=0
s
y
, (3.78)
where
F
M
(3.79)
45
3 Tire
limits the camber force to the maximum tire force. By replacing the partial derivative of the
lateral tire force at a vanishing lateral slip by the global derivative of the generalized tire force
dF
y
s
y
s
y
=0
F
s
(3.80)
the camber force will be automatically reduced on increasing slip, Fig. 3.30.
The camber angle inuences the distribution of pressure in the lateral direction of the contact
patch, and changes the shape of the contact patch from rectangular to trapezoidal. Thus, it is
extremely difcult, if not impossible, to quantify the camber inuence with the aid of such
simple models. But, it turns out that this approach is quit a good approximation.
3.4.8 Bore Torque
3.4.8.1 Modeling Aspects
The angular velocity of the wheel consists of two components
0W
=
0R
+ e
yR
. (3.81)
The wheel rotation itself is represented by e
yR
, whereas
0R
describes the motions of the
knuckle without any parts into the direction of the wheel rotation axis. In particular during
steering motions the angular velocity of the wheel has a component in direction of the track
normal e
n
n
= e
T
n

0W
,= 0 (3.82)
which will cause a bore motion. If the wheel moves in longitudinal and lateral direction too
then, a very complicated deection prole of the tread particles in the contact patch will occur.
However, by a simple approach the resulting bore torque can be approximated quite good by
the parameter of the generalized tire force characteristics.
At rst, the complex shape of a tires contact patch is approximated by a circle, Fig. 3.32.
By setting
R
P
=
1
2
_
L
2
+
B
2
_
=
1
4
(L +B) (3.83)
the radius of the circle can be adjusted to the length L and the width B of the actual contact
patch. During pure bore motions circumferential forces F are generated at each patch element
dA at the radius r. The integration over the contact area A
T
B
=
1
A
_
A
F r dA (3.84)
will then produce the resulting bore torque.
46
normal shape of contact patch
circular
approximation
d
F
R
P
r
dr
n
B
L
e
x
e
y
Figure 3.32: Bore torque approximation
3.4.8.2 Maximum Torque
At large bore motions all particles in the contact patch are sliding. Then, F = F
S
= const. will
hold and Eq. (3.84) simplies to
T
max
B
=
1
A
F
S
_
A
r dA . (3.85)
With dA = r ddr and A = R
2
P
one gets
T
max
B
=
1
R
2
P

F
S
R
P
_
0
2
_
0
r rddr =
2
R
2
P
F
S
R
P
_
0
r
2
dr =
2
3
R
P
F
S
= R
B
F
S
, (3.86)
where
R
B
=
2
3
R
P
(3.87)
can be considered as the bore radius of the contact patch.
3.4.8.3 Bore Slip
For small slip values the force transmitted in the patch element can be approximated by
F = F(s) dF
0
s (3.88)
where s denotes the slip of the patch element, and dF
0
is the initial inclination of the generalized
tire force characteristics. Similar to Eqs. (3.59) and (3.61) we dene
s =
r
n
r
D
[[
(3.89)
47
3 Tire
where r
n
describes the sliding velocity in the patch element and the term r
D
[[ consisting
of the dynamic tire radius r
D
and the angular velocity of the wheel represents the average
transport velocity of the tread particles. By setting r = R
B
we can dene a bore slip now
s
B
=
R
B

n
r
D
[[
. (3.90)
Then, Eq. (3.92) simplies to
s =
r
R
B
s
B
. (3.91)
Inserting Eqs. (3.88) and (3.91) into Eq. (3.84) results in
T
B
= =
1
R
2
P

R
P
_
0
2
_
0
dF
0
r
R
B
s
B
r rddr . (3.92)
As the bore slip s
B
does not depend on r Eq. (3.92) simplies to
T
B
=
2
R
2
P
dF
0
s
B
R
B
R
P
_
0
r
3
dr =
2
R
2
P
dF
0
s
B
R
B
R
4
P
4
=
1
2
R
P
dF
0
R
P
R
B
s
B
. (3.93)
With R
P
=
3
2
R
B
one nally gets
T
B
=
9
8
R
B
dF
0
s
B
. (3.94)
Via the initial inclination dF
0
and the bore radius R
B
the bore torque T
B
automatically takes
the actual tire properties into account.
To avoid numerical problems at a locked wheel, where = 0 will hold, the modied bore slip
s
B
=
R
B

n
r
D
[[ +v
N
(3.95)
can be used for practical applications. Where the small positive velocity v
N
> 0 is added in the
denominator.
3.4.8.4 Model Realisation
With regard to the overall model assumptions Eq. (3.94) can be simplied to
T
B
=
9
8
R
B
dF
0
s
B
R
B
dF
0
s
B
. (3.96)
But, it is limited by Eq. (3.86) for large bore motions. Hence, the simple, but nonlinear bore
torque model nally is given by
T
B
= R
B
dF
0
s
B
with [ T
B
[ R
B
F
S
, (3.97)
where the bore radius R
B
and the bore slip s
B
are dened by Eqs. (3.87) and (3.90) and dF
0
and
F
S
are the initial inclination and the sliding value of the generalized tire force characteristics.
48
3.5 First Order Tire Dynamics
3.5 First Order Tire Dynamics
Measurements show that the dynamic reaction of the tire forces and torques to disturbances can
be approximated quite well by rst order systems. Then, the dynamic tire forces F
D
x
, F
D
y
and
the dynamic tire torque T
D
z
can be modeled by rst order differential equations
x

F
D
x
+ F
D
x
= F
S
x
y

F
D
y
+ F
D
y
= F
S
y

T
D
z
+ T
D
z
= T
S
z
(3.98)
which are driven by the steady values F
S
x
, F
S
y
and T
S
z
. The tread particles of a rolling tire move
with the transport velocity r
D
[[ through the contact patch, where r
D
and denote the dynamic
rolling radius and the angular velocity of the wheel.
Now, time constants
i
, can be derived from so called relaxation lengths r
i
i
=
r
i
r
D
[[
i = x, y, . (3.99)
But it turned out that these relaxation lengths are functions of the longitudinal and lateral slip
s
x
, s
y
and the wheel load F
z
, Fig. 3.33. Therefore, constant relaxation lengths will approximate
1 2 3 4 5 6 8 9 7 10 0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
F
z
= 6 kN
F
z
= 4 kN
F
z
= 2 kN
slip angle [
o
]
r
y
[m]
Figure 3.33: Measured lateral force relaxation length for a typical passenger car tire, [14]
the real tire behavior in zero order approximation only. An appropriate model for the dynamic
tire performance would be of great advantage because then, the cumbersome task of deriving
the relaxation lengths from measurements can be avoided.
A rst order expansion from the steady state characteristics to dynamic tire forces and torques
can be found in [42]. This model approach leads to relaxation lengths which are automatically
adapted to the tire parameter.
49
4 Suspension System
4.1 Purpose and Components
The automotive industry uses different kinds of wheel/axle suspension systems. Important cri-
teria are costs, space requirements, kinematic properties, and compliance attributes.
The main purposes of a vehicle suspension system are
carry the car and its weight,
maintain correct wheel alignment,
control the vehicles direction of travel,
keep the tires in contact with the road,
reduce the effect of shock forces.
Vehicle suspension systems consist of
guiding elements
control arms, links
struts
leaf springs
force elements
coil spring, torsion bar, air spring, leaf spring
anti-roll bar, anti-sway bar or stabilizer
damper
bushings, hydro-mounts
tires.
Tires are air springs that support the total weight of the vehicle. The air spring action of the tire
is very important to the ride quality and safe handling of the vehicle.
50
4.2 Some Examples
4.2 Some Examples
4.2.1 Multi Purpose Systems
The double wishbone suspension, the McPherson suspension and the multi-link suspension are
multi purpose wheel suspension systems, Fig. 4.1.
Figure 4.1: Double wishbone, McPherson and multi-link suspension
They are used as steered front or non steered rear axle suspension systems. These suspension
systems are also suitable for driven axles.
In a McPherson suspension the spring is mounted with an inclination to the strut axis. Thus,
bending torques at the strut, which cause high friction forces, can be reduced.
leaf springs
links
Figure 4.2: Solid axles guided by leaf springs and links
At pickups, trucks, and busses solid axles are used often. They are guided either by leaf springs
or by rigid links, Fig. 4.2. Solid axles tend to tramp on rough roads.
Leaf-spring-guided solid axle suspension systems are very robust. Dry friction between the leafs
leads to locking effects in the suspension. Although the leaf springs provide axle guidance on
some solid axle suspension systems, additional links in longitudinal and lateral direction are
used. Thus, the typical wind-up effect on braking can be avoided.
Solid axles suspended by air springs need at least four links for guidance. In addition to a good
driving comfort air springs allow level control too.
51
4 Suspension System
4.2.2 Specic Systems
The semi-trailing arm, the short-long-arm axle (SLA), and the twist beam axle suspension are
suitable only for non-steered axles, Fig. 4.3.
Figure 4.3: Specic wheel/axles suspension systems
The semi-trailing arm is a simple and cheap design which requires only few space. It is mostly
used for driven rear axles.
The short-long-arm axle design allows a nearly independent layout of longitudinal and lateral
axle motions. It is similar to the central control arm axle suspension, where the trailing arm is
completely rigid and hence, only two lateral links are needed.
The twist beam axle suspension exhibits either a trailing arm or a semi-trailing arm character-
istic. It is used for non driven rear axles only. The twist beam axle provides enough space for
spare tire and fuel tank.
4.3 Steering Systems
4.3.1 Requirements
The steering system must guarantee easy and safe steering of the vehicle. The entirety of the
mechanical transmission devices must be able to cope with all loads and stresses occurring in
operation.
52
4.3 Steering Systems
In order to achieve a good maneuverability a maximum steering angle of approx. 30
must be
provided at the front wheels of passenger cars. Depending on the wheel base, busses and trucks
need maximum steering angles up to 55
at the front wheels.

Recently some companies have started investigations on steer by wire techniques.
4.3.2 Rack and Pinion Steering
Rack-and-pinion is the most common steering system of passenger cars, Fig. 4.4. The rack may
be located either in front of or behind the axle. Firstly, the rotations of the steering wheel
S
steering
box
rack
d
ra
g
lin
k
wheel
and
wheel
body
u
R
1

2
pinion
S
Figure 4.4: Rack and pinion steering
are transformed by the steering box to the rack travel u
R
= u
R
(
S
) and then via the drag links
transmitted to the wheel rotations
1
=
1
(u
R
),
2
=
2
(u
R
). Hence, the overall steering ratio
depends on the ratio of the steering box and on the kinematics of the steering linkage.
4.3.3 Lever Arm Steering System
steering box
drag link 1
L drag link 2
s
te
e
rin
g
le
v
e
r 2
s
t
e
e
r
in
g

le
v
e
r

1
wheel and
wheel body
Figure 4.5: Lever arm steering system
Using a lever arm steering system Fig. 4.5, large steering angles at the wheels are possible. This
steering system is used on trucks with large wheel bases and independent wheel suspension at
the front axle. Here, the steering box can be placed outside of the axle center.
53
4 Suspension System
Firstly, the rotations of the steering wheel
S
are transformed by the steering box to the ro-
tation of the steer levers
L
=
L
(
S
). The drag links transmit this rotation to the wheel
1
=
1
(
L
),
2
=
2
(
L
). Hence, the overall steering ratio again depends on the ratio of the
steering box and on the kinematics of the steering linkage.
4.3.4 Drag Link Steering System
At solid axles the drag link steering system is used, Fig. 4.6. The rotations of the steering wheel
steer box
(90
o
rotated)
drag link
steering link
ste
e
rin
g
le
ve
r
O
1

2
wheel
and
wheel
body
Figure 4.6: Drag link steering system
S
are transformed by the steering box to the rotation of the steering lever arm
L
=
L
(
S
) and
further on to the rotation of the left wheel,
1
=
1
(
L
). The drag link transmits the rotation of
the left wheel to the right wheel,
2
=
2
(
1
). The steering ratio is dened by the ratio of the
steering box and the kinematics of the steering link. Here, the ratio
2
=
2
(
1
) given by the
kinematics of the drag link can be changed separately.
4.3.5 Bus Steer System
In busses the driver sits more than 2 m in front of the front axle. In addition, large steering
angles at the front wheels are needed to achieve a good manoeuvrability. That is why, more
sophisticated steering systems are needed, Fig. 4.7. The rotations of the steering wheel
S
are
transformed by the steering box to the rotation of the steering lever arm
L
=
L
(
S
). The left
lever arm is moved via the steering link
A
=
A
(
L
). This motion is transferred by a coupling
link to the right lever arm. Finally, the left and right wheels are rotated via the drag links,
1
=
1
(
A
) and
2
=
2
(
A
).
54
4.4 Standard Force Elements
steering box
steering link
1
drag link
coupl.
link
le
ft
le
ve
r
a
rm
s
te
e
rin
g
le
v
e
r
A
wheel and
wheel body
L
Figure 4.7: Typical bus steering system
4.4.1 Springs
Springs support the weight of the vehicle. In vehicle suspensions coil springs, air springs, torsion
bars, and leaf springs are used, Fig. 4.8.
Coil spring
Leaf spring
Torsion bar u
u
u
Air spring
u
F
S
F
S
F
S
F
S
Figure 4.8: Vehicle suspension springs
Coil springs, torsion bars, and leaf springs absorb additional load by compressing. Thus, the ride
height depends on the loading condition. Air springs are rubber cylinders lled with compressed
air. They are becoming more popular on passenger cars, light trucks, and heavy trucks because
here the correct vehicle ride height can be maintained regardless of the loading condition by
adjusting the air pressure.
55
4 Suspension System
c
L
F
S
L
F
L
u
F
S
c
L
0
u
F
S
F
S
0
Figure 4.9: Linear coil spring and general spring characteristics
Alinear coil spring may be characterized by its free length L
F
and the spring stiffness c, Fig. 4.9.
The force acting on the spring is then given by
F
S
= c
_
L
F
L
_
, (4.1)
where L denotes the actual length of the spring. Mounted in a vehicle suspension the spring
has to support the corresponding chassis weight. Hence, the spring will be compressed to the
conguration length L
0
< L
F
. Now, Eq. (4.1) can be written as
F
S
= c
_
L
F
(L
0
u)
_
= c
_
L
F
L
0
_
+c u = F
0
S
+c u , (4.2)
where F
0
S
is the spring preload and u describes the spring displacement measured from the
springs conguration length.
In general the spring force F
S
can be dened by a nonlinear function of the spring displacement
u
F
S
= F
S
(u) . (4.3)
Now, arbitrary spring characteristics can be approximated by elementary functions, like poly-
nomials, or by tables which are then inter- and extrapolated by linear functions or cubic splines.
The complex behavior of leaf springs and air springs can only be approximated by simple
nonlinear spring characteristics, F
S
= F
S
(u). For detailed investigations sophisticated, [44] or
even dynamic spring models, [8] have to be used.
4.4.2 Anti-Roll Bar
The anti-roll or anti-sway bar or stabilizer is used to reduce the roll angle during cornering and
to provide additional stability. Usually, it is simply a U-shaped metal rod connected to both
of the lower control arms, Fig. 4.10. Thus, the two wheels of an axle are interconnected by a
torsion bar spring. This affects each one-sided bouncing. The axle with the stronger stabilizer
is rather inclined to breaking out, in order to reduce the roll angle.
56
lo
w
e
r
c
o
n
tr
o
l a
r
m
u
p
p
e
r
c
o
n
tr
o
l a
r
m
s
te
e
r
in
g
b
o
x
a
n
ti-
r
o
ll b
a
r
b
e
a
r
in
g
s
to
c
h
a
s
s
is
lin
k
to
lo
w
e
r
c
o
n
tr
o
l a
r
m
s
2
s
1
Figure 4.10: Axle with anti-roll bar attached to lower control arms
When the suspension at one wheel moves up and on the other down the anti-roll bar generates a
force acting in opposite direction at each wheel. In a good approximation this force is given by
F
arb
= c
arb
(s
1
s
2
) , (4.4)
where s
1
, s
2
denote the vertical suspension motion of the left and right wheel center, and c
W
arb
in
[N/m] names the stiffness of the anti-roll bar with respect to the vertical suspension motions of
the wheel centers.
F
F
z
a
d
b
Fa
Fa
z
1
-z
2
Figure 4.11: Anti-roll bar loaded by vertical forces
Assuming a simple U-shaped anti-roll bar the stiffness of the anti-roll bar is dened by the
geometry and material properties. Vertical forces with the magnitude F applied in opposite di-
rection at both ends to the anti-roll bar, result in the vertical displacement z measured between
both ends of the anti-roll bar, Fig. 4.11. The stiffness of the anti-roll bar itself is then dened by
c =
F
z
. (4.5)
57
4 Suspension System
Neglecting all bending effects one gets
z = a = a
Fa b
G

32
D
4
, (4.6)
where G denotes the modulus of shear and the distances a and b are dened in Fig. 4.11. Hence,
the stiffness of the anti-roll bar is given by
c =

32
G D
4
a
2
b
. (4.7)
Depending on the axle design the ends of the ant-roll bar are attached via links to the knuckle
or, as shown in Fig. refFig:susp Axle with anti-roll bar, to the lower control arm. In both cases
the displacement of the anti-roll bar end is given as a function of the vertical suspension motion
of the wheel center. For small displacements one gets
z
1
= i
arb
s
1
and z
2
= i
arb
s
2
, (4.8)
where i
arb
denotes the ratio of the vertical motions of the wheel centers s
1
, s
2
and the anti-roll
bar ends z
1
, z
2
. Now, the the stiffness of the anti-roll bar with respect to the vertical suspension
motions of the wheel centers is given by
c
arb
= i
2
arb
32
G D
4
a
2
b
. (4.9)
The stiffness strongly depends (forth power) on the diameter of the anti-roll bar.
For a typical passenger car the following data will hold: a = 230mm, b = 725mm, D = 20mm
and i
arb
= 2/3. The shear modulus of steel is given by G = 85 000 N/mm
2
. Then, one gets
c
arb
=
_
2
3
_
2
32
85 000 N/mm
2
(20 mm)
4
(230 mm)
2
725 mm
= 15.5 N/mm = 15 500 N/m. (4.10)
This simple calculation will produce the real stiffness not exactly, because bending effects and
compliancies in the anti-roll bar bearings will reduce the the stiffness of the anti-roll bar.
4.4.3 Damper
Dampers are basically oil pumps, Fig. 4.12. As the suspension travels up and down, the hy-
draulic uid is forced by a piston through tiny holes, called orices. This slows down the sus-
pension motion.
Today twin-tube and mono-tube dampers are used in vehicle suspension systems. Dynamic
damper model, like the one in [1], compute the damper force via the uid pressure applied to
each side of the piston. The change in uid pressures in the compression and rebound chambers
are calculated by applying the conservation of mass.
58
Remote Oil Ch.
Rebound Ch.
Remote Gas Chamber
Compression
Chamber
Piston
F
D
v
F
D
Piston orifice
Remote orifice
Figure 4.12: Principle of a mono-tube damper
In standard vehicle dynamics applications simple characteristics
F
D
= F
D
(v) (4.11)
are used to describe the damper force F
D
as a function of the damper velocity v. To obtain this
characteristics the damper is excited with a sinusoidal displacement signal u = u
0
sin 2ft. By
varying the frequency in several steps from f = f
0
to f = f
E
different force displacement
curves F
D
= F
D
(u) are obtained, Fig. 4.13. By taking the peak values of the damper force at
the displacement u = u
0
which corresponds with the velocity v = 2fu
0
the characteristics
F
D
= F
D
(v) is generated now. Here, the rebound cycle is associated with negative damper
velocities.
1000
0 -0.04 0
-1000
-2000
-3000
-0.4 -0.8 -1.2 -1.6 -0.02 0.02 0.04
0
-4000
-0.06 0.06 0.8 1.6 1.2 0.4
F
D
= F
D
(u) F
D
= F
D
(v)
Rebound
Compression

F
D

[
N
]
u [m]
v [m/s]
f
E
f
0
Figure 4.13: Damper characteristics generated from measurements, [14]
Typical passenger car or truck dampers will have more resistance during its rebound cycle then
its compression cycle.
4.4.4 Rubber Elements
Force elements made of natural rubber or urethane compounds are used in many locations on
the vehicle suspension system, Fig. 4.14. Those elements require no lubrication, isolate minor
59
4 Suspension System
vibration, reduce transmitted road shock, operate noise free, offer high load carrying capabili-
ties, and are very durable.
Control arm
bushings
Subframe mounts
Topmount
Stop
Figure 4.14: Rubber elements in vehicle suspension
During suspension travel, the control arm bushings provide a pivot point for the control arm.
They also maintain the exact wheel alignment by xing the lateral and vertical location of the
control arm pivot points. During suspension travel the rubber portion of the bushing must twist
to allow control arm motion. Thus, an additional resistance to suspension motion is generated.
Bump and rebound stops limit the suspension travel. The compliance of the topmount avoids
the transfer of large shock forces to the chassis. The subframe mounts isolate the suspension
system from the chassis and allow elasto-kinematic steering effects of the whole axle.
It turns out, that those elastic elements can hardly be described by simple spring and damper
characteristics, F
S
= F
S
(u) and F
D
= F
D
(v), because their stiffness and damping properties
change with the frequency of the motion. Here, more sophisticated dynamic models are needed.
4.5 Dynamic Force Elements
4.5.1 Testing and Evaluating Procedures
The effect of dynamic force elements is usually evaluated in the frequency domain. For this, on
test rigs or in simulation the force element is excited by sine waves
x
e
(t) = A sin(2 f t) , (4.12)
with different frequencies f
0
f f
E
and amplitudes A
min
A A
max
. Starting at t = 0,
the system will usually be in a steady state condition after several periods t nT, where T =
1/f and n = 2, 3, . . . have to be chosen appropriately. Due to the nonlinear system behavior the
system response is periodic, F(t +T) = F(T), where T = 1/f, yet not harmonic. That is why,
60
the measured or calculated force F will be approximated within one period nT t (n+1)T,
by harmonic functions as good as possible
F(t)
..
measured/
calculated
sin(2 f t) + cos(2 f t)
. .
rst harmonic approximation
. (4.13)
The coefcients and can be calculated from the demand for a minimal overall error
1
2
(n+1)T
_
nT
_
sin(2 f t)+ cos(2 f t) F(t)
_
2
dt Minimum . (4.14)
The differentiation of Eq. (4.14) with respect to and yields two linear equations as necessary
conditions
(n+1)T
_
nT
_
_
sin(2 f t) dt = 0
(n+1)T
_
nT
_
_
cos(2 f t) dt = 0
(4.15)
with the solutions
=
_
F sin dt
_
cos
2
dt
_
F cos dt
_
sin cos dt
_
sin
2
dt
_
cos
2
dt 2
_
sin cos dt
=
_
F cos dt
_
sin
2
dt
_
F sin dt
_
sin cos dt
_
sin
2
dt
_
cos
2
dt 2
_
sin cos dt
, (4.16)
where the integral limits and arguments of sine and cosine no longer have been written. Because
it is integrated exactly over one period nT t (n + 1)T, for the integrals in Eq. (4.16)
_
sin cos dt = 0 ;
_
sin
2
dt =
T
2
;
_
cos
2
dt =
T
2
(4.17)
hold, and as solution
=
2
T
_
F sin dt , =
2
T
_
F cos dt (4.18)
remains. However, these are exactly the rst two coefcients of a Fourier-"-Approximation.
The rst order harmonic approximation in Eq. (4.13) can now be written as
F(t) =

F sin (2 f t + ) (4.19)
where amplitude

F and phase angle are given by
F =
_
2
+
2
and tan =

. (4.20)
61
4 Suspension System
A simple force element consisting of a linear spring with the stiffness c and a linear damper
with the constant d in parallel would respond with
F(t) = c x
e
+ d x
e
= c A sin 2ft + d 2f A cos 2ft . (4.21)
Here, amplitude and phase angle are given by
F = A
_
c
2
+ (2fd)
2
and tan =
d 2f A
c A
= 2f
d
c
. (4.22)
Hence, the response of a pure spring, c ,= 0 and d = 0 is characterized by

F = Ac and
tan = 0 or = 0, whereas a pure damper response with c = 0 and d ,= 0 results in
F = 2fdA and tan or = 90
. Hence, the phase angle which is also called

the dissipation angle can be used to evaluate the damping properties of the force element. The
dynamic stiffness, dened by
c
dyn
=
F
A
(4.23)
is used to evaluate the stiffness of the element.
In practice the frequency response of a system is not determined punctually, but continuously.
For this, the system is excited by a sweep-sine. In analogy to the simple sine-function
x
e
(t) = A sin(2 f t) , (4.24)
where the period T = 1/f appears as pre-factor at differentiation
x
e
(t) = A2 f cos(2 f t) =
2
T
A cos(2 f t) . (4.25)
A generalized sine-function can be constructed, now. Starting with
x
e
(t) = A sin(2 h(t)) , (4.26)
the time derivative results in
x
e
(t) = A2

h(t) cos(2 h(t)) . (4.27)
In the following we demand that the function h(t) generates periods fading linearly in time, i.e:
h(t) =
1
T(t)
=
1
p q t
, (4.28)
where p > 0 and q > 0 are constants yet to determine. Eq. (4.28) yields
h(t) =
1
q
ln(p q t) + C . (4.29)
The initial condition h(t = 0) = 0 xes the integration constant
C =
1
q
ln p . (4.30)
62
With Eqs. (4.30) and (4.29) Eq. (4.26) results in a sine-like function
x
e
(t) = A sin
_
2
q
ln
p
p q t
_
, (4.31)
which is characterized by linear fading periods.
The important zero values for determining the period duration lie at
1
q
ln
p
p q t
n
= 0, 1, 2, or
p
p q t
n
= e
nq
, mit n = 0, 1, 2, (4.32)
and
t
n
=
p
q
_
1 e
nq
_
, n = 0, 1, 2, . (4.33)
The time difference between two zero values yields the period
T
n
= t
n+1
t
n
=
p
q
_
1e
(n+1) q
1+e
nq
_
T
n
=
p
q
e
nq
_
1 e
q
_
, n = 0, 1, 2, . (4.34)
For the rst (n = 0) and last (n = N) period one nds
T
0
=
p
q
_
1 e
q
_
T
N
=
p
q
_
1 e
q
_
e
N q
= T
0
e
N q
. (4.35)
With the frequency range to investigate, given by the initial f
0
and nal frequency f
E
, the
parameters q and the ratio q/p can be calculated from Eq. (4.35)
q =
1
N
ln
f
E
f
0
,
q
p
= f
0
_
1
_
f
E
f
0
_ 1
N
_
, (4.36)
with N xing the number of frequency intervals. The passing of the whole frequency range then
takes the time
t
N+1
=
1 e
(N+1) q
q/p
. (4.37)
Hence, to test or simulate a force element in the frequency range from 0.1 Hz to f = 100 Hz
with N = 500 intervals will only take 728 s or 12min.
4.5.2 Simple Spring Damper Combination
Fig. 4.15 shows a simple dynamic force element where a linear spring with the stiffness c and a
linear damper with the damping constant d are arranged in series.
The displacements of the force element and the spring itself are described by u and s. Then, the
the forces acting in the spring and damper are given by
F
S
= c s and F
D
= d ( u s) . (4.38)
63
4 Suspension System
s u
c d
Figure 4.15: Spring and damper in series
The force balance F
D
= F
S
results in a linear rst order differential equation for the spring
displacement s
d ( u s) = c s or
d
c
s = s +
d
c
u , (4.39)
where the ratio between the damping coefcient d and the spring stiffness c acts as time con-
stant, T = d/c. Hence, this force element will responds dynamically to any excitation.
The steady state response to a harmonic excitation
u(t) = u
0
sin t respectively u = u
0
cos t (4.40)
can be calculated easily. The steady state response will be of the same type as the excitation.
Inserting
s
(t) = u
0
(a sin t +b cos t) (4.41)
into Eq. (4.39) results in
d
c
u
0
(acos t bsin t)
. .
s
= u
0
(a sin t +b cos t)
. .
s
+
d
c
u
0
cos t
. .
u
. (4.42)
Collecting all sine and cosine terms we obtain two equations
d
c
u
0
b = u
0
a and
d
c
u
0
a = u
0
b +
d
c
u
0
(4.43)
which can be solved for the two unknown parameter
a =

2
2
+ (c/d)
2
and b =
c
d
2
+ (c/d)
2
. (4.44)
Hence, the steady state force response reads as
F
S
= c s
= c u
0
2
+ (c/d)
2
_
sin t +
c
d
cos t
_
(4.45)
which can be transformed to
F
S
=

F
S
sin (t + ) (4.46)
where the force magnitude

F
S
and the phase angle are given by
F
S
=
c u
0

2
+ (c/d)
2
_
2
+ (c/d)
2
=
c u
0

_
2
+ (c/d)
2
and = arctan
c/d
. (4.47)
64
0
100
200
300
400
0 20 40 60 80 100
0
50
100
c
dyn
[N/mm]
[
o
]
f [Hz]
c = 400 N/mm
d
1
= 1000 N/(m/s)
d
2
= 2000 N/(m/s)
d
1
= 3000 N/(m/s)
d
2
= 4000 N/(m/s)
1
1
2
3
4
2
3
4
c
d
Figure 4.16: Frequency response of a spring damper combination
The dynamic stiffness c
dyn
=

F
S
/u
0
and the phase angle are plotted in Fig. 4.16 for different
damping values.
With increasing frequency the spring damper combination changes from a pure damper perfor-
mance, c
dyn
0 and 90
to a pure spring behavior, c

dyn
c and 0. The frequency
range, where the element provides stiffness and damping is controlled by the value for the
damping constant d.
4.5.3 General Dynamic Force Model
To approximate the complex dynamic behavior of bushings and elastic mounts different spring
damper models can be combined. A general dynamic force model is constructed by N parallel
force elements, Fig. 4.17. The static load is carried by a single spring with the stiffness c
0
or an
arbitrary nonlinear force characteristics F
0
= F
0
(u).
Within each force element the spring acts in serial to parallel combination of a damper and a
dry friction element. Now, even hysteresis effects and the stress history of the force element can
be taken into account.
The forces acting in the spring and damper of force element i are given by
F
Si
= c
i
s
i
and F
Di
= d
i
( s
i
u) , (4.48)
were u and s
i
describe the overall element and the spring displacement.
65
4 Suspension System
c
1 c
2
c
N
d
1
s
1
d
2
s
2
d
N
s
N
u
F
F1
F
M
F
F2
F
M
F
FN
F
M
c
0
Figure 4.17: Dynamic force model
As long as the absolute value of the spring force F
Si
is lower than the maximum friction force
F
M
Fi
the damper friction combination will not move at all
u s
i
= 0 for [F
Si
[ F
M
Fi
. (4.49)
In all other cases the force balance
F
Si
= F
Di
F
M
Fi
(4.50)
holds. Using Eq. 4.48 the force balance results in
d
i
( s
i
u) = F
Si
F
M
Fi
(4.51)
which can be combined with Eq. 4.49 to
d
i
s
i
=
_
_
F
Si
+ F
M
Fi
F
Si
<F
M
Fi
d
i
u for F
M
Fi
F
Si
+F
M
Fi
F
Si
F
M
Fi
+F
M
Fi
<F
Si
(4.52)
where according to Eq. 4.48 the spring force is given by F
Si
= c
i
s
i
.
In extension to this linear approach nonlinear springs and dampers may be used. To derive all
the parameters an extensive set of static and dynamic measurements is needed.
4.5.3.1 Hydro-Mount
For the elastic suspension of engines in vehicles very often specially developed hydro-mounts
are used. The dynamic nonlinear behavior of these components guarantees a good acoustic
decoupling but simultaneously provides sufcient damping.
Fig. 4.18 shows the principle and mathematical model of a hydro-mount. At small deformations
the change of volume in chamber 1 is compensated by displacements of the membrane. When
66
main spring
chamber 1
membrane
ring channel
x
e
c
2
T
c
F
M
F
u
F
__ c
2
T __
d
2
F __
d
2
F __
chamber 2
Figure 4.18: Hydro-mount
the membrane reaches the stop, the liquid in chamber 1 is pressed through a ring channel into
chamber 2. The ratio of the chamber cross section to the ring channel cross section is very large.
Thus the uid is moved through the ring channel at very high speed. This results in remarkable
inertia and resistance forces (damping forces).
The force effect of a hydro-mount is combined from the elasticity of the main spring and the
volume change in chamber 1.
With u
F
labeling the displacement of the generalized uid mass M
F
,
F
H
= c
T
x
e
+ F
F
(x
e
u
F
) (4.53)
holds, where the force effect of the main spring has been approximated by a linear spring with
the constant c
T
.
With M
FR
as the actual mass in the ring channel and the cross sections A
K
, A
R
of chamber and
ring channel the generalized uid mass is given by
M
F
=
_
A
K
A
R
_
2
M
FR
. (4.54)
The uid in chamber 1 is not being compressed, unless the membrane can evade no longer. With
the uid stiffness c
F
and the membrane clearance s
F
, one gets
F
F
(x
e
u
F
) =
_
_
c
F
_
(x
e
u
F
) + s
F
_
(x
e
u
F
) < s
F
0 for [x
e
u
f
[ s
F
c
F
_
(x
e
u
F
) s
F
_
(x
e
u
f
) > +s
F
(4.55)
The hard transition from clearance F
F
= 0 and uid compression resp. chamber deformation
with F
F
,= 0 is not realistic and leads to problems, even with the numeric solution. Therefore,
the function (4.55) is smoothed by a parabola in the range [x
e
u
f
[ 2 s
F
.
67
4 Suspension System
The motions of the uid mass cause friction losses in the ring channel, which are as a rst
approximation proportional to the speed,
F
D
= d
F
u
F
. (4.56)
Then, the equation of motion for the uid mass reads as
M
F
u
F
= F
F
F
D
. (4.57)
The membrane clearance makes Eq. (4.57) nonlinear and only solvable by numerical integra-
tion. The nonlinearity also affects the overall force in the hydro-mount, Eq. (4.53).
The dynamic stiffness and the dissipation angle of a hydro-mount are displayed in Fig. 4.19
versus the frequency.
0
100
200
300
400
10
0
10
1
0
10
20
30
40
50
60
Dissipation Angle [deg] at Excitation Amplitudes A = 2.5/0.5/0.1 mm
Excitation Frequency [Hz]
Dynamic Stiffness [N/m] at Excitation Amplitudes A = 2.5/0.5/0.1 mm
Figure 4.19: Dynamic Stiffness [N/mm] and Dissipation Angle [deg] for a Hydro-Mount
The simulation is based on the following system parameters
68
m
F
= 25 kg generalized uid mass
c
T
= 125 000 N/m stiffness of main spring
d
F
= 750 N/(m/s) damping constant
c
F
= 100 000 N/m uid stiffness
s
F
= 0.0002 mm clearance in membrane bearing
By the nonlinear and dynamic behavior a very good compromise can be achieved between noise
isolation and vibration damping.
69
5 Vertical Dynamics
5.1 Goals
The aim of vertical dynamics is the tuning of body suspension and damping to guarantee good
ride comfort, resp. a minimal stress of the load at sufcient safety.
The stress of the load can be judged fairly well by maximal or integral values of the body
accelerations.
The wheel load F
z
is linked to the longitudinal F
x
and lateral force F
y
by the coefcient of
friction. The digressive inuence of F
z
on F
x
and F
y
as well as non-stationary processes at the
increase of F
x
and F
y
in the average lead to lower longitudinal and lateral forces at wheel load
variations.
Maximal driving safety can therefore be achieved with minimal variations of the wheel load.
Small variations of the wheel load also reduce the stress on the track.
The comfort of a vehicle is subjectively judged by the driver. In literature, i.e. [20], different
approaches of describing the human sense of vibrations by different metrics can be found.
Transferred to vehicle vertical dynamics, the driver primarily registers the amplitudes and ac-
celerations of the body vibrations. These values are thus used as objective criteria in practice.
5.2 Basic Tuning
5.2.1 From complex to simple models
For detailed investigations of ride comfort and ride safety sophisticated road and vehicle models
are needed, [45]. The three-dimensional vehicle model, shown in Fig. 5.1, includes an elastically
suspended engine, and dynamic seat models. The elasto-kinematics of the wheel suspension was
described fully nonlinear. In addition, dynamic force elements for the damper elements and the
hydro-mounts are used. Such sophisticated models not only provide simulation results which
are in good conformity to measurements but also make it possible to investigate the vehicle
dynamic attitude in an early design stage.
Much simpler models can be used, however, for basic studies on ride comfort and ride safety.
A two-dimensional vehicle model, for instance, suits perfectly with a single track road model,
Fig. 5.2. Neglecting longitudinal accelerations, the vehicle chassis only performs hub and pitch
motions. Here, the chassis is considered as one rigid body. Then, mass and inertia properties
70
5.2 Basic Tuning
XXXXX X XXX
YYYYY Y YYY
ZZZZZ Z ZZZ
Time = 0.000000
Thilo Seibert Ext. 37598
Vehicle Dynamics, Ford Research Center Aachen
/export/ford/dffa089/u/tseiber1/vedyna/work/results/mview.mvw 07/02/98
AA/FFA
Ford

Figure 5.1: Full Vehicle Model
C
a
1
a
2
M*
M
1
M
2
m
1
m
2
z
A1
z
C1
z
R
(s-a
2
)
z
C2
z
R
(s+a
1
)
z
B
x
B
y
B
C
hub
M,
z
A2
pitch
z
R
(s)
s
Figure 5.2: Vehicle Model for Basic Comfort and Safety Analysis
can be represented by three point masses which are located in the chassis center of gravity and
on top of the front and the rear axle. The lumped mass model has 4 degrees of freedom. The
hub and pitch motion of the chassis are represented by the vertical motions of the chassis in the
front z
C1
and in the rear z
C2
. The coordinates z
A1
and z
A2
describe the vertical motions of the
front and rear axle. The function z
R
(s) provides road irregularities in the space domain, where s
denotes the distance covered by the vehicle and measured at the chassis center of gravity. Then,
the irregularities at the front and rear axle are given by z
R
(s +a
1
) and z
R
(s a
2
) respectively,
where a
1
and a
2
locate the position of the chassis center of gravity C.
The point masses must add up to the chassis mass
M
1
+M
+M
2
= M (5.1)
and they have to provide the same inertia around an axis located in the chassis center C and
pointing into the lateral direction
a
2
1
M
1
+a
2
2
M
2
= . (5.2)
71
5 Vertical Dynamics
The correct location of the center of gravity is assured by
a
1
M
1
= a
2
M
2
. (5.3)
Now, Eqs. (5.2) and (5.3) yield the main masses
M
1
=

a
1
(a
1
+a
2
)
and M
2
=

a
2
(a
1
+a
2
)
, (5.4)
and the coupling mass
M
= M
_
1

Ma
1
a
2
_
(5.5)
follows from Eq. (5.1).
If the mass and the inertia properties of a real vehicle happen to result in = Ma
1
a
2
then,
the coupling mass vanishes M
= 0, and the vehicle can be represented by two uncoupled two

mass systems describing the vertical motion of the axle and the hub motion of the chassis mass
on top of each axle.
vehicles
properties
mid
size
car
full
size
car
sports
utility
vehicle
commercial
vehicle
heavy
truck
front axle
mass
m
1
[kg] 80 100 125 120 600
rear axle
mass
m
2
[kg] 80 100 125 180 1100
center
of
gravity
a
1
[m]
a
2
[m]
1.10
1.40
1.40
1.40
1.45
1.38
1.90
1.40
2.90
1.90
chassis
mass
M [kg] 1100 1400 1950 3200 14300
chassis
inertia
[kg m
2
] 1500 2350 3750 5800 50000
lumped
mass
model
M
1
M
M
2
[kg]
545
126
429
600
200
600
914
76
960
925
1020
1255
3592
5225
5483
Table 5.1: Mass and Inertia Properties of different Vehicles
Depending on the actual mass and inertia properties the vertical dynamics of a vehicle can be
investigated by two simple decoupled mass models describing the vibrations of the front and
rear axle and the corresponding chassis masses. By using half of the chassis and half of the axle
mass we nally end up in quarter car models.
The data in Table 5.1 show that for a wide range of passenger cars the coupling mass is smaller
than the corresponding chassis masses, M
< M
1
and M
< M
2
. Here, the two mass model
or the quarter car model represent a quite good approximation to the lumped mass model. For
72
5.2 Basic Tuning
commercial vehicles and trucks, where the coupling mass has the same magnitude as the corre-
sponding chassis masses, the quarter car model serves for basic studies only.
At most vehicles, c.f. Table 5.1, the axle mass is much smaller than the corresponding chassis
mass, m
i
M
i
, i = 1, 2. Hence, for a rst basic study axle and chassis motions can be in-
vestigated independently. The quarter car model is now further simplied to two single mass
models, Fig. 5.3.
z
R
6
c
c
S
`
`
`
`
`
`
M
d
S
z
C
6
z
R
6
c
T
c
`
`
`
`
`
`
z
W
6 m
c
S
`
`
`
`
`
`
d
S
Figure 5.3: Simple Vertical Vehicle Models
The chassis model neglects the tire deection and the inertia forces of the wheel. For the high
frequent wheel motions the chassis can be considered as xed to the inertia frame.
The equations of motion for the models read as
M z
C
+ d
S
z
C
+ c
S
z
C
= d
S
z
R
+ c
S
z
R
(5.6)
and
m z
W
+ d
S
z
W
+ (c
S
+c
T
) z
W
= c
T
z
R
, (5.7)
where z
C
and z
W
label the vertical motions of the corresponding chassis mass and the wheel
mass with respect to the steady state position. The constants c
S
, d
S
describe the suspension
stiffness and damping. The dynamic wheel load is calculated by
F
D
T
= c
T
(z
R
z
W
) (5.8)
where c
T
is the vertical or radial stiffness of the tire and z
R
denotes the road irregularities. In
this simple approach the damping effects in the tire are not taken into account.
5.2.2 Natural Frequency and Damping Rate
At an ideally even track the right side of the equations of motion (5.6), (5.7) vanishes because
of z
R
=0 and z
R
=0. The remaining homogeneous second order differential equations can be
written in a more general form as
z + 2
0
z +
2
0
z = 0 , (5.9)
73
5 Vertical Dynamics
where
0
represents the undamped natural frequency, and is a dimensionless parameter called
viscous damping ratio. For the chassis and the wheel model the new parameter are given by
Chassis: z z
C
,
C
=
d
S
2
c
S
M
,
2
0

2
0C
=
c
S
M
;
Wheel: z z
W
,
W
=
d
S
2
_
(c
S
+c
T
)m
,
2
0

2
0W
=
c
S
+c
T
m
.
(5.10)
The solution of Eq. (5.9) is of the type
z(t) = z
0
e
t
, (5.11)
where z
0
and are constants. Inserting Eq. (5.11) into Eq. (5.9) results in
(
2
+ 2
0
+
2
0
) z
0
e
t
= 0 . (5.12)
Non-trivial solutions z
0
,= 0 are possible, if
2
+ 2
0
+
2
0
= 0 (5.13)
will hold. The roots of the characteristic equation (5.13) depend on the value of
< 1 :
1,2
=
0
i
0
_
1
2
,
1 :
1,2
=
0
_

_
2
1
_
.
(5.14)
Figure 5.4 shows the root locus of the eigenvalues for different values of the viscous damping
rate .
-3 -2.5 -2 -1.5
-1
-0.5
-1.0
-0.5
0
0.5
1.0
Re()
/
0
Im()
/
0
=1
=1.25 =1.25 =1.5
=1.5
=0
=0
=0.2
=0.2
=0.5
=0.5
=0.7
=0.7
=0.9
=0.9
Figure 5.4: Eigenvalues
1
and
2
for different values of
74
5.2 Basic Tuning
For 1 the eigenvalues
1,2
are both real and negative. Hence, Eq. (5.11) will produce a
exponentially decaying solution. If < 1 holds, the eigenvalues
1,2
will become complex,
where
2
is the complex conjugate of
1
. Now, the solution can be written as
z(t) = Ae
0
t
sin
_
0
_
1
2
t
_
, (5.15)
where Aand are constants which have to be adjusted to given initial conditions z(0) = z
0
and
z(0) = z
0
. The real part Re (
1,2
) =
0
is negative and determines the decay of the solution.
The imaginary Im(
1,2
) =
0
_
1
2
part denes the actual frequency of the vibration. The
actual frequency
=
0
_
1
2
(5.16)
tends to zero, 0, if the viscous damping ratio will approach the value one, 1.
In a more general way the relative damping may be judged by the ratio
D
=
Re(
1,2
)
[
1,2
[
. (5.17)
For complex eigenvalues which characterize vibrations
D
= (5.18)
holds, because the absolute value of the complex eigenvalues is given by
[
1,2
[ =
_
Re(
1,2
)
2
+Im(
1,2
)
2
=
_
(
0
)
2
+
_
0
_
1
2
_
2
=
0
, (5.19)
and hence, Eq. (5.17) results in
D
=
+
0
0
= . (5.20)
For 1 the eigenvalues become real and negative. Then, Eq. (5.17) will always produce the
relative damping value D
= 1. In this case the viscous damping rate is more sensitive.

5.2.3 Spring Rates
5.2.3.1 Minimum Spring Rates
The suspension spring is loaded with the corresponding vehicle weight. At linear spring char-
acteristics the steady state spring deection is calculated from
u
0
=
M g
c
S
. (5.21)
At a conventional suspension without niveau regulation a load variation M M +M leads
to changed spring deections u
0
u
0
+ u. In analogy to (5.21) the additional deection
follows from
u =
M g
c
S
. (5.22)
75
5 Vertical Dynamics
If for the maximum load variation M the additional spring deection is limited to u the
suspension spring rate can be estimated by a lower bound
c
S

M g
u
. (5.23)
In the standard design of a passenger car the engine is located in the front and the trunk in the
rear part of the vehicle. Hence, most of the load is supported by the rear axle suspension.
For an example we assume that 150 kg of the permissible load of 500 kg are going to the front
axle. Then, each front wheel is loaded by M
F
= 150 kg/2 = 75 kg and each rear wheel by
M
R
= (500 150) kg/2 = 175 kg.
The maximum wheel travel is limited, u u
max
. At standard passenger cars it is in the range
of u
max
0.8 m to u
max
0.10 m. By setting u = u
max
/2 we demand that the spring
deection caused by the load should not exceed half of the maximum value. Then, according to
Eq. (5.23) a lower bound of the spring rate at the front axle can be estimated by
c
min
S
= ( 75 kg 9.81 m/s
2
)/(0.08/2) m = 18400 N/m. (5.24)
The maximum load over one rear wheel was estimated here by M
R
= 175 kg. Assuming that
the suspension travel at the rear axle is slightly larger, u
max
0.10 m the minimum spring rate
at the rear axle can be estimated by
c
min
S
= ( 175 kg 9.81 m/s
2
)/(0.10/2) m = 34300 N/m, (5.25)
which is nearly two times the minimum value of the spring rate at the front axle. In order to
reduce this difference we will choose a spring rate of c
S
= 20 000 N/m at the front axle.
In Tab. 5.1 the lumped mass chassis model of a full size passenger car is described by M
1
=
M
2
= 600 kg and M
= 200. To approximate the lumped mass model by two decoupled two

mass models we have to neglect the coupling mass or, in order to achieve the same chassis
mass, to distribute M
equally to the front and the rear. Then, the corresponding cassis mass of
a quarter car model is given here by
M =
_
M
1
+M
/2
_
/2 = (600 kg + 200/2 kg)/2 = 350 kg . (5.26)
According to Eq. 5.10 the undamped natural eigen frequency of the simple chassis model is
then given by
2
0C
= c
S
/M. Hence, for a spring rate of c
S
= 20000 N/m the undamped natural
frequency of the unloaded car amounts to
f
0C
=
_
20000 N/m 350 kg/(2 ) = 1.2 Hz , (5.27)
which is a typical value for most of all passenger cars. Due to the small amount of load the
undamped natural frequency for the loaded car does not change very much,
f
0C
=
_
20000 N/m (350 + 75) kg/(2 ) = 1.1 Hz . (5.28)
The corresponding cassis mass over the rear axle is given here by
M =
_
M
2
+M
/2
_
/2 = (600 kg + 200/2 kg)/2 = 350 kg . (5.29)
76
5.2 Basic Tuning
Now the undamped natural frequencies for the unloaded
f
0
0C
=
_
34300 N/m/350 kg/(2 ) = 1.6 Hz (5.30)
and the loaded car
f
L
0C
=
_
34300 N/m/(350 + 175) kg/(2 ) = 1.3 Hz (5.31)
are larger and differ more.
5.2.3.2 Nonlinear Springs
In order to reduce the spring rate at the rear axle and to avoid too large spring deections when
loaded nonlinear spring characteristics are used, Fig. 5.5. Adding soft bump stops the overall
spring force in the compression mode u 0 can be modeled by the nonlinear function
F
S
= F
0
S
+ c
0
u
_
1 +k
_
u
u
_
2
_
, (5.32)
where F
0
S
is the spring preload, c
S
describes the spring rate at u = 0, and k > 0 characterizes the
intensity of the nonlinearity. The linear characteristic provides at u = u the value F
lin
S
(u) =
F
0
S
+c
S
u. To achieve the same value with the nonlinear spring
F
0
S
+c
0
u (1 +k) = F
0
S
+c
S
u or c
0
(1 +k) = c
S
(5.33)
must hold, where c
S
describes the spring rate of the corresponding linear characteristics. The
local spring rate is determined by the derivative
dF
S
du
= c
0
_
1 + 3 k
_
u
u
_
2
_
. (5.34)
Hence, the spring rate for the loaded car at u = u is given by
c
L
= c
0
(1 + 3 k) . (5.35)
The intensity of the nonlinearity k can be xed, for instance, by choosing an appropriate spring
rate for the unloaded vehicle. With c
0
= 20000 N/m the spring rates on the front and rear axle
will be the same for the unloaded vehicle. With c
S
= 34300 N/m Eq. (5.33) yields
k =
c
S
c
0
1 =
34300
20000
1 = 0.715 . (5.36)
The solid line in Fig. 5.5 shows the resulting nonlinear spring characteristics which is charac-
terized by the spring rates c
0
= 20 000 N/m and c
L
= c
0
(1 +3k) = 20 000 (1 +3 0.715) =
62 900 N/m for the unloaded and the loaded vehicle. Again, the undamped natural frequencies
f
0
0C
=
20000 N/m
350 kg
1
2
= 1.20 Hz or f
L
0C
=
92000 N/m
(350+175) kg
1
2
= 1.74 Hz (5.37)
77
5 Vertical Dynamics
F
S
u
F
S
0
u
M g
dF
S
du
u=u
dF
S
du
u=0
c
S
F
S

[
N
]
u
[
m
]
63 kN/m
44 kN/m
20 kN/m
29 kN/m
0 0.05 0.1
2000
4000
6000
8000
Figure 5.5: Principle and realizations of nonlinear spring characteristics
for the unloaded and the loaded vehicle differ quite a lot.
The unloaded and the loaded vehicle have the same undamped natural frequencies if
c
0
M
=
c
L
M +M
or
c
L
c
0
=
M +M
M
(5.38)
will hold. Combing this relationship with Eq. (5.35) one obtains
1 + 3 k =
M
M +M
or k =
1
3
_
M +M
M
1
_
=
1
3
M
M
. (5.39)
Hence, for the quarter car model with M = 350kg and M = 175 the intensity of the nonlinear
spring amounts to k = 1/3 175/350 = 0.1667. This value and c
S
= 34300 N/m will produce
the dotted line in Fig. 5.5. The spring rates c
0
= c
S
/(1 + k) = 34 300 N/m/ (1 + 0.1667) =
29 400N/mfor the unloaded and c
L
= c
0
(1+3k) = 29 400N/m(1+30.1667) = 44 100N/m
for the loaded vehicle follow from Eqs. (5.34) and (5.35). Now, the undamped natural frequency
for the unloaded f
0
0C
=
_
c
0
/M = 1.46 Hz and the loaded vehicle f
0
0C
=
_
c
L
/(M +M) =
1.46 Hz are in deed the same.
5.2.4 Inuence of Damping
To investigate the inuence of the suspension damping to the chassis and wheel motion the
simple vehicle models are exposed to initial disturbances. Fig. 5.6 shows the time response of
the chassis z
C
(t) and wheel displacement z
W
(t) as well as the chassis acceleration z
C
and the
wheel load F
T
= F
0
T
+ F
D
T
for different damping rates
C
and
W
. The dynamic wheel load
follows from Eq. (5.8), and the static wheel load is given by F
0
T
= (M + m) g, where g labels
the constant of gravity.
To achieve the same damping rates for the chassis and the wheel model different values for the
damping parameter d
S
were needed.
78
5.2 Basic Tuning
0 0.5 1 1.5
-100
-50
0
50
100
150
200
displacement [mm]
t [s]
C
[ - ] d
S
[Ns/m]
0.25
0.50
0.75
1.00
1.25
1323
2646
3969
5292
6614
0 0.05 0.1 0.15
0
1000
2000
3000
4000
5000
6000
50 kg
20000 N/m
220000 N/m
d
S
wheel load [N]
t [s]
20000 N/m
350 kg
d
S
W
[ - ] d
S
[Ns/m]
0.25
0.50
0.75
1.00
1.25
1732
3464
5196
6928
8660
-1
-0.5
0
0.5
1
0 0.5 1 1.5
t [s]
acceleration [g]
0 0.05 0.1 0.15
-10
-5
0
5
10
15
20
chassis model wheel model
displacement [mm]
t [s]
C

W
W
Figure 5.6: Time response of simple vehicle models to initial disturbances
With increased damping the overshoot effect in the time history of the chassis displacement and
the wheel load becomes smaller and smaller till it vanishes completely at
C
= 1 and
W
= 1.
The viscous damping rate = 1
5.2.5 Optimal Damping
5.2.5.1 Avoiding Overshoots
If avoiding overshoot effects is the design goal then, = 1 will be the optimal damping ratio.
For = 1 the eigenvalues of the single mass oscillator change from complex to real. Thus,
producing a non oscillating solution without any sine and cosine terms.
79
5 Vertical Dynamics
According to Eq. (5.10)
C
= 1 and
W
= 1 results in the optimal damping parameter
d
opt
S
C
=1
Comfort
= 2
_
c
S
M , and d
opt
S
W
=1
Safety
= 2
_
(c
S
+c
T
)m . (5.40)
So, the damping values
d
opt
S
C
=1
Comfort
= 5292
N
m/s
and d
opt
S
W
=1
Safety
= 6928
N
m/s
(5.41)
will avoid an overshoot effect in the time history of the chassis displacement z
C
(t) or in the in
the time history of the wheel load F
T
(t). Usually, as it is here, the damping values for optimal
comfort and optimal ride safety will be different. Hence, a simple linear damper can either avoid
overshoots in the chassis motions or in the wheel loads.
The overshot in the time history of the chassis accelerations z
C
(t) will only vanish for
C

which surely is not a desirable conguration, because then, it takes a very long time till the
initial chassis displacement has fully disappeared.
5.2.5.2 Disturbance Reaction Problem
Instead of avoiding overshoot effects we better demand that the time history of the system
response will approach the steady state value as fast as possible. Fig. 5.7 shows the typical
time response of a damped single-mass oscillator to the initial disturbance z(t = 0) = z
0
and
z(t =0) = 0.
z(t)
t
z
0
t
E
z
S
Figure 5.7: Evaluating a damped vibration
Counting the differences of the system response z(t) from the steady state value z
S
= 0 as
errors allows to judge the attenuation. If the overall quadratic error becomes a minimum
2
=
t=t
E
_
t=0
z(t)
2
dt Min , (5.42)
the system approaches the steady state position as fast as possible. In theory t
E
holds, for
practical applications a nite t
E
have to be chosen appropriately.
To judge ride comfort and ride safety the hub motion of the chassis z
C
, its acceleration z
C
and
the variations of the dynamic wheel load F
D
T
can be used. In the absence of road irregularities
80
5.2 Basic Tuning
z
R
= 0 the dynamic wheel load from Eq. (5.8) simplies to F
D
T
= c
T
z
W
. Hence, the demands
2
C
=
t=t
E
_
t=0
_
_
g
1
z
C
_
2
+
_
g
2
z
C
_
2
_
dt Min (5.43)
and
2
S
=
t=t
E
_
t=0
_
c
T
z
W
_
2
dt Min (5.44)
will guarantee optimal ride comfort and optimal ride safety. By the factors g
1
and g
2
the accel-
eration and the hub motion can be weighted differently.
The equation of motion for the chassis model can be resolved for the acceleration
z
C
=
_
2
0C
z
C
+ 2
C
z
C
_
, (5.45)
where, the system parameter M, d
S
and c
S
were substituted by the damping rate
C
=
C

0C
=
d
S
/(2M) and by the undamped natural frequency
0C
= c
S
/M. Then, the problemin Eq. (5.43)
can be written as
2
C
=
t=t
E
_
t=0
_
g
2
1
_
2
0C
z
C
+ 2
C
z
C
_
2
+ g
2
2
z
2
C
_
dt
=
t=t
E
_
t=0
_
z
C
z
C
_
_
g
2
1
(
2
0C
)
2
+g
2
2
g
2
1

2
0C
2
C
g
2
1

2
0C
2
C
g
2
1
(2
C
)
2
_
_
_
_
z
C
z
C
_
_
Min ,
(5.46)
where x
T
C
=
_
z
C
z
C
is the state vector of the chassis model. In a similar way Eq. (5.44) can
be transformed to
2
S
=
t=t
E
_
t=0
c
2
T
z
2
W
dt =
t=t
E
_
t=0
_
z
W
z
W
_
c
2
T
0
0 0
__
z
W
z
W
_
Min , (5.47)
where x
T
W
=
_
z
W
z
W
denotes the state vector of the wheel model.

The problems given in Eqs. (5.46) and (5.47) are called disturbance-reaction problems, [7].
They can be written in a more general form
t=t
E
_
t=0
x
T
(t) Qx(t) dt Min (5.48)
where x(t) denotes the state vector and Q = Q
T
is a symmetric weighting matrix. For single
mass oscillators described by Eq. (5.9) the state equation reads as
_
z
z
_
. .
x
=
_
0 1
2
0
2
_
. .
A
_
z
z
_
. .
x
. (5.49)
81
5 Vertical Dynamics
For t
E
the time response of the system exposed to the initial disturbance x(t =0) = x
0
vanishes x(t ) = 0, and the integral in Eq.(5.48) can be solved by
t=t
E
_
t=0
x
T
(t) Qx(t) dt = x
T
0
Rx
0
, (5.50)
where the symmetric matrix R = R
T
is given by the Ljapunov equation
A
T
R + RA + Q = 0 . (5.51)
For the single mass oscillator the Ljapunov equation
_
0
2
0
1 2
_ _
R
11
R
12
R
12
R
22
_
+
_
R
11
R
12
R
12
R
22
_ _
0 1
2
0
2
_
+
_
Q
11
Q
12
Q
12
Q
22
_
. (5.52)
results in 3 linear equations
2
0
R
12
2
0
R
12
+ Q
11
= 0
2
0
R
22
+R
11
2 R
12
+ Q
12
= 0
R
12
2 R
22
+R
12
2 R
22
+ Q
22
= 0
(5.53)
which easily can be solved for the elements of R
R
11
=
_

2
0
+
1
4
_
Q
11
Q
12
+

2
0
4
Q
22
, R
12
=
Q
11
2
2
0
, R
22
=
Q
11
4
2
0
+
Q
22
4
. (5.54)
For the initial disturbance x
0
= [ z
0
0 ]
T
Eq. (5.50) nally results in
t=t
E
_
t=0
x
T
(t) Qx(t) dt = z
2
0
R
11
= z
2
0
__

2
0
+
1
4
_
Q
11
Q
12
+

2
0
4
Q
22
_
. (5.55)
Now, the integral in Eq. (5.46) evaluating the ride comfort is solved by
2
C
= z
2
0C
__

C
2
0C
+
1
4
C
_
_
g
2
1
_
2
0C
_
2
+g
2
2
_
g
2
1

2
0C
2
C
+

2
0C
4
C
g
2
1
(2
C
)
2
_
= z
2
0C

2
0C
_
0C
4
C
_
g
2
1
+
_
g
2
2
0C
_
2
_
+
_
g
2
2
0C
_
2
C

0C
_
.
(5.56)
where the abbreviation
C
was nally replaced by
C

0C
.
By setting g
1
= 1 and g
2
= 0 the time history of the chassis acceleration z
C
is weighted only.
Eq. (5.56) then simplies to
2
C
z
C
= z
2
0C

2
0C
0C
4
C
(5.57)
82
5.2 Basic Tuning
which will become a minimum for
0C
0 or
C
. As mentioned before,
C

surely is not a desirable conguration. A low undamped natural frequency
0C
0 is achieved
by a soft suspension spring c
S
0 or a large chassis mass M . However, a large chas-
sis mass is uneconomic and the suspension stiffness is limited by the the loading conditions.
Hence, weighting the chassis accelerations only does not lead to a specic result for the system
parameter.
The combination of g
1
= 0 and g
2
= 1 weights the time history of the chassis displacement
only. Then, Eq. (5.56) results in
2
C
z
C
=
z
2
0C
0C
_
1
4
C
+
C
_
(5.58)
which will become a minimum for
0C
or
d
2
C
[
z
C
d
C
=
z
2
0C
0C
_
1
4
2
C
+ 1
_
= 0 . (5.59)
A high undamped natural frequency
0C
contradicts the demand for rapidly vanishing
accelerations. The viscous damping ratio
C
=
1
2
solves Eq. (5.59) and minimizes the merit
function in Eq. (5.58). But again, this value does not correspond with
C
which minimizes
the merit function in Eq. (5.57).
Hence, practical results can be achieved only if the chassis displacements and the chassis ac-
celerations will be evaluated simultaneously. To do so, appropriate weighting factors have to
be chosen. In the equation of motion for the chassis (5.6) the terms M z
C
and c
S
z
C
are added.
Hence, g
1
= M and g
2
= c
S
or
g
1
= 1 and g
2
=
c
S
M
=
2
0C
(5.60)
provide system-tted weighting factors. Now, Eq. (5.56) reads as
2
C
= z
2
0C

2
0C
_
0C
2
C
+
C

0C
_
. (5.61)
Again, a good ride comfort will be achieved by
0C
0. For nite undamped natural frequen-
cies Eq. (5.61) becomes a minimum, if the viscous damping rate
C
will satisfy
d
2
C
[
z
C
d
C
= z
2
0C

2
0C
_
0C
2
2
C
+
0C
_
= 0 . (5.62)
Hence, a viscous damping rate of
C
=
1
2
2 (5.63)
or a damping parameter of
d
opt
S
C
=
1
2
2
Comfort
=
_
2 c
S
M , (5.64)
will provide optimal comfort by minimizing the merit function in Eq. (5.61).
83
5 Vertical Dynamics
For the passenger car with M = 350 kg and c
S
= 20 000 N/m the optimal damping parameter
will amount to
d
opt
S
C
=
1
2
2
Comfort
= 3742
N
m/s
(5.65)
which is 70% of the value needed to avoid overshot effects in the chassis displacements.
The integral in Eq. (5.47) evaluating the ride safety is solved by
2
S
=
z
2
0W
0W
_
W
+
1
4
W
_
c
2
T
(5.66)
where the model parameter m, c
S
, d
S
and c
T
where replaced by the undamped natural frequency
2
0W
= (c
S
+c
T
)/m and by the damping ratio
W
=
W

0W
= d
S
/(2m).
A soft tire c
T
0 make the safety criteria Eq. (5.66) small
2
S
0 and thus, reduces the
dynamic wheel load variations. However, the tire spring stiffness can not be reduced to arbitrary
low values, because this would cause too large tire deformations. Small wheel masses m 0
and/or a hard body suspension c
S
will increase
0W
and thus, reduce the safety criteria
Eq. (5.66). The use of light metal rims improves, because of wheel weight reduction, the ride
safety of a car. Hard body suspensions contradict a good driving comfort.
With xed values for c
T
and
0W
the merit function in Eq. (5.66) will become a minimum if
2
S
W
=
z
2
0W
0W
_
1 +
1
4
2
W
_
c
2
T
= 0 (5.67)
will hold. Hence, a viscous damping rate of
W
=
1
2
(5.68)
or the damping parameter
d
opt
S
Safety
=
_
(c
S
+c
T
) m (5.69)
will guarantee optimal ride safety by minimizing the merit function in Eq. (5.66).
For the passenger car with M = 350 kg and c
S
= 20 000 N/m the optimal damping parameter
will now amount to
d
opt
S
W
=
1
2
Safety
= 3464
N
m/s
(5.70)
which is 50% of the value needed to avoid overshot effects in the wheel loads.
5.3 Sky Hook Damper
5.3.1 Modelling Aspects
In standard vehicle suspension systems the damper is mounted between the wheel and the body.
Hence, the damper affects body and wheel/axle motions simultaneously.
84
5.3 Sky Hook Damper
d
S
c
S
c
T
M
m
z
C
z
W
z
R
sky
d
W
d
B
c
S
c
T
M
m
z
C
z
W
z
R
F
D
a) Standard Damper b) Sky Hook Damper
Figure 5.8: Quarter Car Model with Standard and Sky Hook Damper
To take this situation into account the simple quarter car models of section 5.2.1 must be com-
bined to a more enhanced model, Fig. 5.8a.
Assuming a linear characteristics the suspension damper force is given by
F
D
= d
S
( z
W
z
C
) , (5.71)
where d
S
denotes the damping constant, and z
C
, z
W
are the time derivatives of the absolute
vertical body and wheel displacements.
The sky hook damping concept starts with two independent dampers for the body and the
wheel/axle mass, Fig. 5.8b. A practical realization in form of a controllable damper will then
provide the damping force
F
D
= d
W
z
W
d
C
z
C
, (5.72)
where instead of the single damping constant d
S
now two design parameter d
W
and d
C
are
available.
The equations of motion for the quarter car model are given by
M z
C
= F
S
+F
D
M g ,
m z
W
= F
T
F
S
F
D
mg ,
(5.73)
where M, m are the sprung and unsprung mass, z
C
, z
W
denote their vertical displacements, and
g is the constant of gravity.
The suspension spring force is modeled by
F
S
= F
0
S
+ c
S
(z
W
z
C
) , (5.74)
where F
0
S
= m
C
g is the spring preload, and c
S
the spring stiffness.
85
5 Vertical Dynamics
Finally, the vertical tire force is given by
F
T
= F
0
T
+ c
T
(z
R
z
W
) , (5.75)
where F
0
T
= (M + m) g is the tire preload, c
S
the vertical tire stiffness, and z
R
describes the
road roughness. The condition F
T
0 takes the tire lift off into account.
5.3.2 Eigenfrequencies and Damping Ratios
Using the force denitions in Eqs. (5.72), (5.74) and (5.75) the equations of motion in Eq. (5.73)
can be transformed to the state equation
_
_
z
C
z
W
z
C
z
W
_
_
. .
x
=
_
_
0 0 1 0
0 0 0 1
c
S
M
c
S
M

d
C
M
d
W
M
c
S
m

c
S
+c
T
m
d
C
m

d
W
m
_
_
. .
A
_
_
z
C
z
W
z
C
z
W
_
_
. .
x
+
_
_
0
0
0
c
T
m
_
_
. .
B
_
z
R
_
. .
u
, (5.76)
where the weight forces Mg, mg were compensated by the preloads F
0
S
, F
0
T
, the term Bu
describes the excitation, x denotes the state vector, and A is the state matrix. In this linear
approach the tire lift off is no longer taken into consideration.
The eigenvalues of the state matrix A will characterize the eigen dynamics of the quarter car
model. In case of complex eigenvalues the damped natural eigenfrequencies are given by the
imaginary parts, = Im(), and according to Eq. (vdyn-eq: relative damping ratio lambda)
= D
= Re()/ [[. evaluates the damping ratio.

By setting d
C
= d
S
and d
W
= d
S
Eq. (5.76) represents a quarter car model with the standard
damper described by Eq. (5.71). Fig. 5.9 shows the eigenfrequencies f = /(2) and the
damping ratios = D
for different values of the damping parameter d

S
.
Optimal ride comfort with a damping ratio of
C
=
1
2
2 0.7 for the chassis motion could

be achieved with the damping parameter d
S
= 3880 N/(m/s), and the damping parameter
d
S
= 3220 N/(m/s) would provide for the wheel motion a damping ratio of
W
= 0.5 which
correspond to minimal wheel load variations. This damping parameter are very close to the val-
ues 3742 N/(m/s) and 3464 N/(m/s) which very calculated in Eqs. (5.65) and (5.70) with the
single mass models. Hence, the very simple single mass models can be used for a rst damper
layout. Usually, as it is here, optimal ride comfort and optimal ride safety cannot achieved both
by a standard linear damper.
The sky-hook damper, modeled by Eq. (5.72), provides with d
W
and d
S
two design parameter.
Their inuence to the eigenfrequencies f and the damping ratios is shown in Fig. 5.10.
The the sky-hook damping parameter d
C
, d
W
have a nearly independent inuence on the damp-
ing ratios. The chassis damping ratio
C
mainly depends on d
C
, and the wheel damping ratio
W
mainly depends on d
W
. Hence, the damping of the chassis and the wheel motion can be
adjusted to nearly each design goal. Here, a sky-hook damper with d
C
= 3900 N/(m/s) and
d
W
= 3200 N/(m/s) would generate the damping ratios d
C
= 0.7 and d
W
= 0.5 hence,
combining ride comfort and ride safety within one damper layout.
86
5.3 Sky Hook Damper
0 1000 2000 3000 4000 5000
0
0.2
0.4
0.6
0.8
1
0 1000 2000 3000 4000 5000
0
2
4
6
8
10
12
Damping ratio = D
Frequencies [Hz]
d
S
[N/(m/s)] d
S
[N/(m/s)]
Chassis
Wheel
0.7
0.5
3220
3880
350 kg
50 kg
d
S
220000 N/m
20000 N/m
Figure 5.9: Quarter car model with standard damper
0 1000 2000 3000 4000 5000
0
2
4
6
8
10
12
0 1000 2000 3000 4000 5000
0
0.2
0.4
0.6
0.8
1
350 kg
50 kg
d
C
220000 N/m
20000 N/m
d
W
d
C
4500
4000
3500
3000
2500
2000
1500
d
C
[N/(m/s)]
d
W
[N/(m/s)] d
W
[N/(m/s)]
Damping ratios
C
,
W
Frequencies [Hz]
0.7
0.5
C
Figure 5.10: Quarter car model with sky-hook damper
5.3.3 Technical Realization
By modifying the damper law in Eq. (5.72) to
F
D
= d
W
z
W
d
C
z
C
+ =
d
W
z
W
d
C
z
C
z
W
z
C
. .
d
S
( z
W
z
C
) = d
S
( z
W
z
C
) (5.77)
the sky-hook damper can be realized by a standard damper in the form of Eq. (5.71). The
new damping parameter d
S
now nonlinearly depends on the absolute vertical velocities of the
chassis and the wheel d
S
= d
S
( z
C
, z
W
). As, a standard damper operates in a dissipative mode
only the damping parameter will be restricted to positive values, d
S
> 0. Hence, the passive
realization of a sky-hook damper will only match with some properties of the ideal damper law
in Eq. (5.72). But, compared with the standard damper it still can provide a better ride comfort
combined with an increased ride safety.
87
5 Vertical Dynamics
5.4 Nonlinear Force Elements
5.4.1 Quarter Car Model
The principal inuence of nonlinear characteristics on driving comfort and safety can already
be studied on a quarter car model Fig. 5.11.
c
T
m
M
nonlinear damper nonlinear spring
z
C
z
W
z
R
F
D
v
u
F
S
F
S
F
D
Figure 5.11: Quarter car model with nonlinear spring and damper characteristics
The equations of motion read as
M z
C
= F
S
+F
D
M g
m z
W
= F
T
F
S
F
D
mg ,
(5.78)
where g = 9.81 m/s
2
labels the constant of gravity, M, m are the masses of the chassis and the
wheel, F
S
, F
D
, F
T
describe the spring, the damper, and the vertical tire force, and the vertical
displacements of the chassis z
C
and the wheel z
W
are measured from the equilibrium position.
In extension to Eq. (5.32) the spring characteristics is modeled by
F
S
= F
0
S
+
_
_
c
0
u
_
1 +k
r
_
u
u
r
_
2
_
u < 0
c
0
u
_
1 +k
c
_
u
u
c
_
2
_
u 0
(5.79)
where F
0
S
= M g is the spring preload, and
u = z
W
z
C
(5.80)
describes the spring travel. Here, u < 0 marks tension (rebound), and u 0 compression. Two
sets of k
r
, u
r
and k
c
, u
c
dene the spring nonlinearity during rebound and compression. For
k
r
= 0 and k
c
= 0 a linear spring characteristics is obtained.
88
A degressive damper characteristics can be modeled by
F
D
(v) =
_
_
d
0
v
1 p
r
v
v < 0 ,
d
0
v
1 + p
c
v
v 0 ,
(5.81)
where d
0
denotes the damping constant at v = 0, and the damper velocity is dened by
v = z
W
z
C
. (5.82)
The sign convention of the damper velocity was chosen consistent to the spring travel. Hence,
rebound is characterized by v < 0 and compression by v 0. The parameter p
r
and p
c
make it
possible to model the damper nonlinearity differently in the rebound and compression mode. A
linear damper characteristics is obtained with p
r
= 0 and p
c
= 0.
The nonlinear spring design in Section 5.2.3 holds for the compression mode. Hence, using
the same data we obtain: c
0
= 29 400 N/m, u
c
= u = u
max
/2 = 0.10/2 = 0.05 and
k
c
= k = 0.1667. By setting u
r
= u
c
and k
r
= 0 a simple linear is used in the rebound mode,
Fig. 5.12a.
0
1000
2000
3000
4000
5000
6000
7000
-5000
-2500
0
2500
5000
-0.05 -0.1 0.05 0 0.1 -0.5 -1 0.5 0 1
c
S
= 34300 N/m
c
0
= 29400 N/m
u
r
= 0.05 m
k
r
= 0
u
c
= 0.05 m
k
c
= 0.1667
u [m]
F
S

[
N
/
m
]
compression
u > 0
rebound
u < 0
compression
v > 0
rebound
v < 0
d
0
= 4200 N/(m/s)
p
r
= 0.4 1/(m/s)
p
c
= 1.2 1/(m/s)
v [m/s]
F
D

[
N
/
m
]
a) Spring b) Damper
Figure 5.12: Spring and damper characteristics: - - - linear, nonlinear
According to Section 5.2.5 damping coefcients optimizing the ride comfort and the ride safety
can be calculated from
Eqs. (5.64) and (5.69). For c
S
= 34 300 N/m which is the equivalent linear spring rate, M =
350 kg, m = 50 kg and c
T
= 220 000 N/m we obtain
(d
S
)
C
opt
=
2 c
S
M =
2 34 300 350 = 4900 N/(m/s) ,

(d
S
)
S
opt
=
_
(c
S
+c
T
) m =
_
(18 000 + 220 000) 50 = 3570 N/(m/s) .
(5.83)
89
5 Vertical Dynamics
The mean value d
0
= 4200 N/(m/s) may serve as compromise. With p
r
= 0.4 (m/s)
1
and p
c
= 1.2 (m/s)
1
the nonlinearity becomes more intensive in compression than rebound,
Fig. 5.12b.
5.4.2 Results
The quarter car model is driven with constant velocity over a single obstacle. Here, a cosine
shaped bump with a height of H = 0.08 m and a length of L = 2.0 m was used. The results are
plotted in Fig. 5.13.
0 0.5 1 1.5
-15
-10
-5
0
5
10
0 0.5 1 1.5
0
1000
2000
3000
4000
5000
6000
7000
0 0.5 1 1.5
-0.06
-0.04
-0.02
0
0.02
0.04
Chassis acceleration [m/s
2
] Wheel load [N]
Suspension travel [m]
time [s] time [s]
time [s]
linear
nonlinear
6660
6160
7.1
6.0
Figure 5.13: Quarter car model driving with v = 20 kmh over a single obstacle
0 0.5 1 1.5 0 0.5 1 1.5 0 0.5 1 1.5
Chassis acceleration [m/s
2
] Wheel load [N]
Suspension travel [m]
time [s] time [s]
time [s]
-15
-10
-5
0
5
10
0
1000
2000
3000
4000
5000
6000
7000
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
linear,
low damping
nonlinear
Figure 5.14: Results for low damping compared to nonlinear model
Compared to the linear model the nonlinear spring and damper characteristics result in signi-
cantly reduced peak values for the chassis acceleration (6.0 m/s
2
instead of 7.1 m/s
2
) and for
90
the wheel load (6160 N instead of 6660 N). Even the tire lift off at t 0.25 s can be avoided.
While crossing the bump large damper velocities occur. Here, the degressive damper charac-
teristics provides less damping compared to the linear damper which increases the suspension
travel.
A linear damper with a lower damping coefcient, d
0
= 3000 N/m for instance, also reduces
the peaks in the chassis acceleration and in the wheel load, but then the attenuation of the
disturbances will take more time. Fig. 5.14. Which surely is not optimal.
91
6 Longitudinal Dynamics
6.1 Dynamic Wheel Loads
6.1.1 Simple Vehicle Model
The vehicle is considered as one rigid body which moves along an ideally even and horizontal
road. At each axle the forces in the wheel contact points are combined in one normal and one
longitudinal force.
S
h
1
a
2
a
mg
v
F
x1
F
x2
F
z2
F
z1
Figure 6.1: Simple vehicle model
If aerodynamic forces (drag, positive and negative lift) are neglected at rst, the equations of
motions in the x-, z-plane will read as
m v = F
x1
+F
x2
, (6.1)
0 = F
z1
+F
z2
mg , (6.2)
0 = F
z1
a
1
F
z2
a
2
+ (F
x1
+F
x2
) h , (6.3)
where v indicates the vehicles acceleration, m is the mass of the vehicle, a
1
+a
2
is the wheel
base, and h is the height of the center of gravity.
These are only three equations for the four unknown forces F
x1
, F
x2
, F
z1
, F
z2
. But, if we insert
Eq. (6.1) in Eq. (6.3), we can eliminate two unknowns at a stroke
0 = F
z1
a
1
F
z2
a
2
+m v h . (6.4)
92
6.1 Dynamic Wheel Loads
The equations Eqs. (6.2) and (6.4) can be resolved for the axle loads now
F
z1
= mg
a
2
a
1
+a
2
h
a
1
+a
2
m v , (6.5)
F
z2
= mg
a
1
a
1
+a
2
+
h
a
1
+a
2
m v . (6.6)
The static parts
F
st
z1
= mg
a
2
a
1
+a
2
, F
st
z2
= mg
a
1
a
1
+a
2
(6.7)
describe the weight distribution according to the horizontal position of the center of gravity. The
height of the center of gravity only inuences the dynamic part of the axle loads,
F
dyn
z1
= mg
h
a
1
+a
2
v
g
, F
dyn
z2
= +mg
h
a
1
+a
2
v
g
. (6.8)
When accelerating v >0, the front axle is relieved as the rear axle is when decelerating v <0.
6.1.2 Inuence of Grade
mg
a
1
a
2
F
x
1
F
z
1
F
x
2
F
z
2
h
v
z
x
Figure 6.2: Vehicle on grade
For a vehicle on a grade, Fig.6.2, the equations of motion Eq. (6.1) to Eq. (6.3) can easily be
extended to
m v = F
x1
+F
x2
mg sin ,
0 = F
z1
+F
z2
mg cos ,
0 = F
z1
a
1
F
z2
a
2
+ (F
x1
+F
x2
) h ,
(6.9)
93
where denotes the grade angle. Now, the axle loads are given by
F
z1
= mg cos
a
2
htan
a
1
+a
2
h
a
1
+a
2
m v , (6.10)
F
z2
= mg cos
a
1
+htan
a
1
+a
2
+
h
a
1
+a
2
m v , (6.11)
where the dynamic parts remain unchanged, whereas now the static parts also depend on the
grade angle and the height of the center of gravity.
6.1.3 Aerodynamic Forces
The shape of most vehicles or specic wings mounted at the vehicle produce aerodynamic
forces and torques. The effect of these aerodynamic forces and torques can be represented by
a resistant force applied at the center of gravity and down forces acting at the front and rear
axle, Fig. 6.3.
mg
a
1
h
a
2
F
D1
F
AR
F
D2
F
x1
F
x2
F
z1 F
z2
Figure 6.3: Vehicle with aerodynamic forces
If we assume a positive driving speed, v > 0, the equations of motion will read as
m v = F
x1
+F
x2
F
AR
,
0 = F
z1
F
D1
+F
z2
F
D2
mg ,
0 = (F
z1
F
D1
) a
1
(F
z2
F
D2
) a
2
+ (F
x1
+F
x2
) h ,
(6.12)
where F
AR
and F
D1
, F
D2
describe the air resistance and the down forces. For the dynamic axle
loads we get
F
z1
= F
D1
+mg
a
2
a
1
+a
2
h
a
1
+a
2
(m v +F
AR
) , (6.13)
F
z2
= F
D2
+mg
a
1
a
1
+a
2
+
h
a
1
+a
2
(m v +F
AR
) . (6.14)
The down forces F
D1
, F
D2
increase the static axle loads, and the air resistance F
AR
generates
an additional dynamic term.
94
6.2 Maximum Acceleration
6.2 Maximum Acceleration
6.2.1 Tilting Limits
Ordinary automotive vehicles can only apply pressure forces to the road. If we take the demands
F
z1
0 and F
z2
0 into account, Eqs. (6.10) and (6.11) will result in
v
g

a
2
h
cos sin and
v
g

a
1
h
cos sin . (6.15)
These two conditions can be combined in one
a
1
h
cos
v
g
+ sin
a
2
h
cos . (6.16)
Hence, the maximum achievable accelerations ( v > 0) and decelerations ( v < 0) are limited
by the grade angle and the position a
1
, a
2
, h of the center of gravity. For v 0 the tilting
condition Eq. (6.16) results in
a
1
h
tan
a
2
h
(6.17)
which describes the climbing and downhill capacity of a vehicle.
The presence of aerodynamic forces complicates the tilting condition. Aerodynamic forces be-
come important only at high speeds. Here, the vehicle acceleration is normally limited by the
engine power.
6.2.2 Friction Limits
The maximum acceleration is also restricted by the friction conditions
[F
x1
[ F
z1
and [F
x2
[ F
z2
(6.18)
where the same friction coefcient has been assumed at front and rear axle. In the limit case
F
x1
= F
z1
and F
x2
= F
z2
(6.19)
the linear momentum in Eq. (6.9) can be written as
m v
max
= (F
z1
+F
z2
) mg sin . (6.20)
Using Eqs. (6.10) and (6.11) one obtains
_
v
g
_
max
= cos sin . (6.21)
That means climbing ( v > 0, > 0) or downhill stopping ( v < 0, < 0) requires at least a
friction coefcient tan [[.
95
According to the vehicle dimensions and the friction values the maximal acceleration or decel-
eration is restricted either by Eq. (6.16) or by Eq. (6.21).
If we take aerodynamic forces into account, the maximum acceleration and deceleration on a
horizontal road will be limited by

_
1 +
F
D1
mg
+
F
D2
mg
_
F
AR
mg

v
g

_
1 +
F
D1
mg
+
F
D2
mg
_
F
AR
mg
. (6.22)
In particular the aerodynamic forces enhance the braking performance of the vehicle.
6.3 Driving and Braking
6.3.1 Single Axle Drive
With the rear axle driven in limit situations, F
x1
=0 and F
x2
=F
z2
hold. Then, using Eq. (6.6)
the linear momentum Eq. (6.1) results in
m v
RWD
= mg
_
a
1
a
1
+a
2
+
h
a
1
+a
2
v
RWD
g
_
, (6.23)
where the subscript
RWD
indicates the rear wheel drive. Hence, the maximum acceleration for
a rear wheel driven vehicle is given by
v
RWD
g
=

1
h
a
1
+a
2
a
1
a
1
+a
2
. (6.24)
By setting F
x1
=F
z1
and F
x2
=0, the maximum acceleration for a front wheel driven vehicle
can be calculated in a similar way. One gets
v
F WD
g
=

1 +
h
a
1
+a
2
a
2
a
1
+a
2
, (6.25)
where the subscript
F WD
denotes front wheel drive. Depending on the parameter , a
1
, a
2
and
h the accelerations may be limited by the tilting condition
v
g

a
2
h
.
The maximum accelerations of a single axle driven vehicle are plotted in Fig. 6.4. For rear
wheel driven passenger cars, the parameter a
2
/(a
1
+a
2
) which describes the static axle load
distribution is in the range of 0.4 a
2
/(a
1
+a
2
) 0.5. For = 1 and h = 0.55 this results
in maximum accelerations in between 0.77 v/g 0.64. Front wheel driven passenger cars
usually cover the range 0.55 a
2
/(a
1
+a
2
) 0.60 which produces accelerations in the range
of 0.45 v/g 0.49. Hence, rear wheel driven vehicles can accelerate much faster than front
wheel driven vehicles.
96
0 0.2 0.4 0.6 0.8 1
0
0.2
0.4
0.6
0.8
1
a
2 /
(a
1
+a
2
)
RWD
FWD
range of load distribution
v

/

g
.
F
W
D
R
W
D
Figure 6.4: Single axle driven passenger car: = 1, h = 0.55 m, a
1
+a
2
= 2.5 m
6.3.2 Braking at Single Axle
If only the front axle is braked, in the limit case F
x1
= F
z1
and F
x2
= 0 will hold. With
Eq. (6.5) one gets from Eq. (6.1)
m v
F WB
= mg
_
a
2
a
1
+a
2
h
a
1
+a
2
v
F WB
g
_
, (6.26)
where the subscript
F WB
indicates front wheel braking. Then, the maximum deceleration is
given by
v
F WB
g
=

1
h
a
1
+a
2
a
2
a
1
+a
2
. (6.27)
If only the rear axle is braked (F
x1
=0, F
x2
=F
z2
), one will obtain the maximum decelera-
tion
v
RWB
g
=

1 +
h
a
1
+a
2
a
1
a
1
+a
2
, (6.28)
where the subscript
RWB
denotes a braked rear axle. Depending on the parameters , a
1
, a
2
,
and h, the decelerations may be limited by the tilting condition
v
g

a
1
h
.
The maximumdecelerations of a single axle braked vehicle are plotted in Fig. 6.5. For passenger
cars the load distribution parameter a
2
/(a
1
+a
2
) usually covers the range of 0.4 to 0.6. If only
the front axle is braked, decelerations from v/g = 0.51 to v/g = 0.77 will be achieved.
This is a quite large value compared to the deceleration range of a braked rear axle which is in
the range of v/g = 0.49 to v/g = 0.33. Therefore, the braking system at the front axle has
a redundant design.
97
0 0.2 0.4 0.6 0.8 1
-1
-0.8
-0.6
-0.4
-0.2
0
a
2 /
(a
1
+a
2
)
range of
load
distribution
v

/

g
.
FWB
RWB
Figure 6.5: Single axle braked passenger car: = 1, h = 0.55 m, a
1
+a
2
= 2.5 m
6.3.3 Braking Stability
On a locked wheel the tire friction force points into the opposite direction of sliding velocity.
Hence, no lateral guidance is provided. That is why a braked vehicle with locked front wheels
will remain stable whereas a braked vehicle with locked rear wheels will become unstable.
v
v
v
F
1
F
2
M
34
locked front wheels
F
3
r
D
3
F
4 r
D
4
v
v
M
12
2
=0
1
=0
Figure 6.6: Locked front wheels
A small yaw disturbance of the vehicle will cause slip angles at the wheels. If the front wheels
are locked and the rear wheels are still rotating, lateral forces can be generated at this axle which
will produce a stabilizing torque, Fig. 6.6. However, a de-stabilizing torque will be generated,
if the rear wheels are locked and the front wheels are still rotating, Fig. 6.7.
98
v
v
4
F
4
locked rear wheels
yaw
disturbance
de-stabilizing
torque
v
x1
F
x1
v
y1
F
y1
v
3
F
3
v
x2
F
x2
v
y2
F
y2
T
12
Figure 6.7: Locked rear wheels
6.3.4 Optimal Distribution of Drive and Brake Forces
The sum of the longitudinal forces accelerates or decelerates the vehicle. In dimensionless style
Eq. (6.1) reads as
v
g
=
F
x1
mg
+
F
x2
mg
. (6.29)
A certain acceleration or deceleration can only be achieved by different combinations of the
longitudinal forces F
x1
and F
x2
. According to Eq. (6.19) the longitudinal forces are limited by
wheel load and friction.
The optimal combination of F
x1
and F
x2
will be achieved, when front and rear axle have the
same skid resistance.
F
x1
= F
z1
and F
x2
= F
z2
. (6.30)
With Eq. (6.5) and Eq. (6.6) one obtains
F
x1
mg
=
_
a
2
h

v
g
_
h
a
1
+a
2
(6.31)
and
F
x2
mg
=
_
a
1
h
+
v
g
_
h
a
1
+a
2
. (6.32)
With Eq. (6.31) and Eq. (6.32) one gets from Eq. (6.29)
v
g
= , (6.33)
where it has been assumed that F
x1
and F
x2
have the same sign. Finally, if Eq. (6.33 is inserted
in Eqs. (6.31) and (6.32) one will obtain
F
x1
mg
=
v
g
_
a
2
h

v
g
_
h
a
1
+a
2
(6.34)
99
and
F
x2
mg
=
v
g
_
a
1
h
+
v
g
_
h
a
1
+a
2
. (6.35)
Depending on the desired acceleration v > 0 or deceleration v < 0, the longitudinal forces that
grant the same skid resistance at both axles can be calculated now.
h=0.55 1
2
-2
-1 0
dF
x2
0
a
1
=1.15
a
2
=1.35
=1.20
a
2
/
h
-a
1
/h
F
x1
/mg
braking
tilting limits
d
r
i
v
i
n
g
dF
x1
F
x
2
/
m
g
B
1
/mg
B
2
/
m
g
Figure 6.8: Optimal distribution of driving and braking forces
Fig. 6.8 shows the curve of optimal drive and brake forces for typical passenger car values. At
the tilting limits v/g = a
1
/h and v/g = +a
2
/h, no longitudinal forces can be applied at the
lifting axle. The initial gradient only depends on the steady state distribution of the wheel loads.
From Eqs. (6.34) and (6.35) it follows
d
F
x1
mg
d
v
g
=
_
a
2
h
2
v
g
_
h
a
1
+a
2
(6.36)
and
d
F
x2
mg
d
v
g
=
_
a
1
h
+ 2
v
g
_
h
a
1
+a
2
. (6.37)
100
For v/g = 0 the initial gradient remains as
d F
x2
d F
x1
0
=
a
1
a
2
. (6.38)
6.3.5 Different Distributions of Brake Forces
Practical applications aim at approximating the optimal distribution of brake forces by constant
distribution, limitation, or reduction of brake forces as good as possible. Fig. 6.9.
F
x1
/mg
F
x
2
/
m
g
constant
distribution
F
x1
/mg
F
x
2
/
m
g
limitation
reduction
F
x1
/mg
F
x
2
/
m
g
Figure 6.9: Different Distributions of Brake Forces
When braking, the stability of a vehicle depends on the potential of generating a lateral force at
the rear axle. Thus, a greater skid (locking) resistance is realized at the rear axle than at the front
axle. Therefore, the brake force distribution are all below the optimal curve in the physically
relevant area. This restricts the achievable deceleration, specially at low friction values.
Because the optimal curve depends on the center of gravity of the vehicle an additional safety
margin have to be installed when designing real brake force distributions. The distribution of
brake forces is often tted to the axle loads. There, the inuence of the height of the cen-
ter of gravity, which may also vary much on trucks, is not taken into account and has to be
compensated by a safety margin from the optimal curve. Only the control of brake pressure in
anti-lock-systems provides an optimal distribution of brake forces independently from loading
conditions.
6.3.6 Anti-Lock-Systems
On hard braking maneuvers large longitudinal slip values occur. Then, the stability and/or steer-
ability is no longer given because nearly no lateral forces can be generated. By controlling the
brake torque or brake pressure respectively, the longitudinal slip can be restricted to values that
allow considerable lateral forces.
Here, the angular wheel acceleration

is used as a control variable. Angular accelerations of
the wheel are derived from the measured angular speeds of the wheel by differentiation. The
rolling condition is fullled with a longitudinal slip of s
L
= 0. Then
r
D

= x (6.39)
holds, where r
D
labels the dynamic tire radius and x names the longitudinal acceleration of the
vehicle. According to Eq. (6.21), the maximum acceleration/deceleration of a vehicle depends
101
on the friction coefcient, [ x[ = g. For a given friction coefcient a simple control law can
be realized for each wheel
[
[
1
r
D
[ x[ . (6.40)
Because no reliable possibility to determine the local friction coefcient between tire and road
has been found until today, useful information can only be gained from Eq. (6.40) at optimal
conditions on dry road. Therefore, the longitudinal slip is used as a second control variable.
In order to calculate longitudinal slips, a reference speed is estimated from all measured wheel
speeds which is used for the calculation of slip at all wheels, then. This method is too imprecise
at low speeds. Therefore, no control is applied below a limit velocity. Problems also arise when
all wheels lock simultaneously for example which may happen on icy roads.
The control of the brake torque is done via the brake pressure which can be increased, held, or
decreased by a three-way valve. To prevent vibrations, the decrement is usually made slower
than the increment.
To prevent a strong yaw reaction, the select low principle is often used with -split braking
at the rear axle. Here, the break pressure at both wheels is controlled by the wheel running on
lower friction. Thus, at least the brake forces at the rear axle cause no yaw torque. However, the
maximum achievable deceleration is reduced by this.
6.4 Drive and Brake Pitch
6.4.1 Vehicle Model
The vehicle model in Fig. 6.10 consists of ve rigid bodies. The body has three degrees of
freedom: Longitudinal motion x
A
, vertical motion z
A
and pitch
A
. The coordinates z
1
and z
2
describe the vertical motions of wheel and axle bodies relative to the body. The longitudinal
and rotational motions of the wheel bodies relative to the body can be described via suspension
kinematics as functions of the vertical wheel motion:
x
1
= x
1
(z
1
) ,
1
=
1
(z
1
) ;
x
2
= x
2
(z
2
) ,
2
=
2
(z
2
) .
(6.41)
The rotation angles
R1
and
R2
describe the wheel rotations relative to the wheel bodies.
The forces between wheel body and vehicle body are labeled F
F1
and F
F2
. At the wheels drive
torques M
A1
, M
A2
and brake torques M
B1
, M
B2
, longitudinal forces F
x1
, F
x2
and the wheel
loads F
z1
, F
z2
apply. The brake torques are directly supported by the wheel bodies, whereas the
drive torques are transmitted by the drive shafts to the vehicle body. The forces and torques that
apply to the single bodies are listed in the last column of the tables 6.1 and 6.2.
The velocity of the vehicle body and its angular velocity are given by
v
0A,0
=
_
_
x
A
0
0
_
_
+
_
_
0
0
z
A
_
_
;
0A,0
=
_
_
0
A
0
_
_
. (6.42)
102
R2
R1
M
B1
M
A1
M
B2
M
A2
A
x
A
z
A
M
B1
M
B2
M
A1
M
A2
z
2
z
1
F
F2
F
F1
F
z1
F
x1
F
z2
F
x2
a
1
R
a
2
h
R
Figure 6.10: Plane Vehicle Model
At small rotational motions of the body one gets for the velocities of the wheel bodies and
wheels
v
0RK
1
,0
= v
0R
1
,0
=
_
_
x
A
0
0
_
_
+
_
_
0
0
z
A
_
_
+
_
_
h
R

A
0
a
1

A
_
_
+
_
_
x
1
z
1
z
1
0
z
1
_
_
; (6.43)
v
0RK
2
,0
= v
0R
2
,0
=
_
_
x
A
0
0
_
_
+
_
_
0
0
z
A
_
_
+
_
_
h
R

A
0
+a
2

A
_
_
+
_
_
x
2
z
2
z
2
0
z
2
_
_
. (6.44)
The angular velocities of the wheel bodies and wheels are obtained from
0RK
1
,0
=
_
_
0
A
0
_
_
+
_
_
0
1
0
_
_
and
0R
1
,0
=
_
_
0
A
0
_
_
+
_
_
0
1
0
_
_
+
_
_
0

R1
0
_
_
(6.45)
as well as
0RK
2
,0
=
_
_
0
A
0
_
_
+
_
_
0
2
0
_
_
and
0R
2
,0
=
_
_
0
A
0
_
_
+
_
_
0
2
0
_
_
+
_
_
0

R2
0
_
_
(6.46)
Introducing a vector of generalized velocities
z =
_
x
A
z
A

1

R1

2

R2
T
, (6.47)
103
the velocities and angular velocities given by Eqs. (6.42), (6.43), (6.44), (6.45), and (6.46) can
be written as
v
0i
=
7
j=1
v
0i
z
j
z
j
and
0i
=
7
j=1
0i
z
j
z
j
(6.48)
6.4.2 Equations of Motion
The partial velocities
v
0i
z
j
and partial angular velocities

0i
z
j
for the ve bodies i =1(1)5 and for
the seven generalized speeds j =1(1)7 are arranged in the tables 6.1 and 6.2.
partial velocities v
0i
/z
j
applied forces
bodies x
A
z
A

A
z
1

R1
z
2

R2
F
e
i
chassis
m
A
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
F
F1
+F
F2
m
A
g
wheel body
front
m
RK1
1
0
0
0
0
1
h
R
0
a
1
x
1
z
1
0
1
0
0
0
0
0
0
0
0
0
0
0
F
F1
m
RK1
g
wheel
front
m
R1
1
0
0
0
0
1
h
R
0
a
1
x
1
z
1
0
1
0
0
0
0
0
0
0
0
0
F
x1
0
F
z1
m
R1
g
wheel body
rear
m
RK2
1
0
0
0
0
1
h
R
0
a
2
0
0
0
0
0
0
x
2
z
2
0
1
0
0
0
0
0
F
F2
m
RK2
g
wheel
rear
m
R2
1
0
0
0
0
1
h
R
0
a
2
0
0
0
0
0
0
x
2
z
2
0
1
0
0
0
F
x2
0
F
z2
m
R2
g
Table 6.1: Partial velocities and applied forces
With the aid of the partial velocities and partial angular velocities the elements of the mass ma-
trix M and the components of the vector of generalized forces and torques Q can be calculated.
M(i, j) =
5
k=1
_
v
0k
z
i
_
T
m
k
v
0k
z
j
+
5
k=1
_
0k
z
i
_
T
0k
z
j
; i, j = 1(1)7 ; (6.49)
Q(i) =
5
k=1
_
v
0k
z
i
_
T
F
e
k
+
5
k=1
_
0k
z
i
_
T
M
e
k
; i = 1(1)7 . (6.50)
Then, the equations of motion for the plane vehicle model are given by
M z = Q. (6.51)
104
partial angular velocities
0i
/z
j
applied torques
bodies x
A
z
A

A
z
1

R1
z
2

R2
M
e
i
chassis
A
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
M
A1
M
A2
a
1
F
F1
+a
2
F
F2
0
wheel body
front
RK1
0
0
0
0
0
0
0
1
0
0
1
z
1
0
0
0
0
0
0
0
0
0
0
0
M
B1
0
wheel
front
R1
0
0
0
0
0
0
0
1
0
0
1
z
1
0
0
1
0
0
0
0
0
0
0
0
M
A1
M
B1
RF
x1
0
wheel body
rear
RK2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
2
z
2
0
0
0
0
0
M
B2
0
wheel
rear
R2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
2
z
2
0
0
1
0
0
M
A2
M
B2
RF
x2
0
Table 6.2: Partial angular velocities and applied torques
6.4.3 Equilibrium
With the abbreviations
m
1
= m
RK1
+m
R1
; m
2
= m
RK2
+m
R2
; m
G
= m
A
+m
1
+m
2
(6.52)
and
h = h
R
+ R (6.53)
The components of the vector of generalized forces and torques read as
Q(1) = F
x1
+F
x2
;
Q(2) = F
z1
+F
z2
m
G
g ;
Q(3) = a
1
F
z1
+a
2
F
z2
h(F
x1
+F
x2
) +a
1
m
1
g a
2
m
2
g ;
(6.54)
Q(4) = F
z1
F
F1
+
x
1
z
1
F
x1
m
1
g +

1
z
1
(M
A1
RF
x1
) ;
Q(5) = M
A1
M
B1
RF
x1
;
(6.55)
Q(6) = F
z2
F
F2
+
x
2
z
2
F
x2
m
2
g +

2
z
2
(M
A
2
RF
x2
) ;
Q(7) = M
A
2
M
B
2
RF
x2
.
(6.56)
Without drive and brake forces
M
A1
= 0 ; M
A2
= 0 ; M
B1
= 0 ; M
B2
= 0 (6.57)
105
from Eqs. (6.54), (6.55) and (6.56) one gets the steady state longitudinal forces, the spring
preloads, and the wheel loads
F
0
x1
= 0 ; F
0
x2
= 0 ;
F
0
F1
=
b
a+b
m
A
g ; F
0
F2
=
a
a+b
m
A
g ;
F
0
z1
= m
1
g +
a
2
a
1
+a
2
m
A
g ; F
0
z2
= m
2
g +
a
1
a
1
+a
2
m
A
g .
(6.58)
6.4.4 Driving and Braking
Assuming that on accelerating or decelerating the vehicle the wheels neither slip nor lock,
R
R1
= x
A
h
R

A
+
x
1
z
1
z
1
;
R
R2
= x
A
h
R

A
+
x
2
z
2
z
2
(6.59)
hold. In steady state the pitch motion of the body and the vertical motion of the wheels reach
constant values
A
=
st
A
= const. , z
1
= z
st
1
= const. , z
2
= z
st
2
= const. (6.60)
and Eq. (6.59) simplies to
R
R1
= x
A
; R
R2
= x
A
. (6.61)
With Eqs. (6.60), (6.61) and (6.53) the equation of motion (6.51) results in
m
G
x
A
= F
a
x1
+F
a
x2
;
0 = F
a
z1
+F
a
z2
;
h
R
(m
1
+m
2
) x
A
+
R1
x
A
R
+
R2
x
A
R
= a F
a
z1
+b F
a
z2
(h
R
+R)(F
a
x1
+F
a
x2
) ;
(6.62)
x
1
z
1
m
1
x
A
+

1
z
1
R1
x
A
R
= F
a
z1
F
a
F1
+
x
1
z
1
F
a
x1
+

1
z
1
(M
A1
RF
a
x1
) ;
R1
x
A
R
= M
A1
M
B1
RF
a
x1
;
(6.63)
x
2
z
2
m
2
x
A
+

2
z
2
R2
x
A
R
= F
a
z2
F
a
F2
+
x
2
z
2
F
a
x2
+

2
z
2
(M
A
2
RF
a
x2
) ;
R2
x
A
R
= M
A
2
M
B
2
RF
a
x2
;
(6.64)
where the steady state spring forces, longitudinal forces, and wheel loads have been separated
into initial and acceleration-dependent terms
F
st
xi
= F
0
xi
+F
a
xi
; F
st
zi
= F
0
zi
+F
a
zi
; F
st
Fi
= F
0
Fi
+F
a
Fi
; i =1, 2 . (6.65)
With given torques of drive and brake the vehicle acceleration x
A
, the wheel forces F
a
x1
, F
a
x2
,
F
a
z1
, F
a
z2
and the spring forces F
a
F1
, F
a
F2
can be calculated from Eqs. (6.62), (6.63) and (6.64).
106
Via the spring characteristics which have been assumed as linear the acceleration-dependent
forces also cause a vertical displacement and pitch motion of the body besides the vertical
motions of the wheels,
F
a
F1
= c
A1
z
a
1
,
F
a
F2
= c
A2
z
a
2
,
F
a
z1
= c
R1
(z
a
A
a
a
A
+z
a
1
) ,
F
a
z2
= c
R2
(z
a
A
+b
a
A
+z
a
2
) .
(6.66)
Especially the pitch of the vehicle
a
A
,=0, caused by drive or brake will be felt as annoying, if
too distinct.
By an axle kinematics with anti dive and/or anti squat properties, the drive and/or brake pitch
angle can be reduced by rotating the wheel body and moving the wheel center in longitudinal
direction during the suspension travel.
6.4.5 Brake Pitch Pole
For real suspension systems the brake pitch pole can be calculated from the motions of the
wheel contact points in the x-, z-plane, Fig. 6.11.
x-, z- motion of the contact points
during compression and rebound
pitch pole
Figure 6.11: Brake Pitch Pole
Increasing the pitch pole height above the track level means a decrease in the brake pitch angle.
However, the pitch pole is not set above the height of the center of gravity in practice, because
the front of the vehicle would rise at braking then.
107
7 Lateral Dynamics
7.1 Kinematic Approach
7.1.1 Kinematic Tire Model
When a vehicle drives through a curve at low lateral acceleration, small lateral forces will be
needed for course holding. Then, hardly lateral slip occurs at the wheels. In the ideal case at van-
ishing lateral slip the wheels only move in circumferential direction. The velocity component
of the contact point in the lateral direction of the tire vanishes then
v
y
= e
T
y
v
0P
= 0 . (7.1)
This constraint equation can be used as kinematic tire model for course calculation of vehicles
moving in the low lateral acceleration range.
7.1.2 Ackermann Geometry
Within the validity limits of the kinematic tire model the necessary steering angle of the front
wheels can be constructed via the given momentary pivot pole M, Fig. 7.1.
At slowly moving vehicles the lay out of the steering linkage is usually done according to the
Ackermann geometry. Then, the following relations apply
tan
1
=
a
R
and tan
2
=
a
R +s
, (7.2)
where s labels the track width and a denotes the wheel base. Eliminating the curve radius R,
we get
tan
2
=
a
a
tan
1
+s
or tan
2
=
a tan
1
a +s tan
1
. (7.3)
The deviations
2
=
a
2

A
2
of the actual steering angle
a
2
from the Ackermann steering
angle
A
2
, which follows from Eq. (7.3), are used, especially on commercial vehicles, to judge
the quality of a steering system.
At a rotation around the momentary pole M, the direction of the velocity is xed for every point
of the vehicle. The angle between the velocity vector v and the longitudinal axis of the vehicle
is called side slip angle. The side slip angle at point P is given by
tan
P
=
x
R
or tan
P
=
x
a
tan
1
, (7.4)
where x denes the distance of P to the inner rear wheel.
108
M
v
2
R
a
s
1

2
x
P
P
Figure 7.1: Ackermann steering geometry at a two-axled vehicle
7.1.3 Space Requirement
The Ackermann approach can also be used to calculate the space requirement of a vehicle
during cornering, Fig. 7.2. If the front wheels of a two-axled vehicle are steered according to
the Ackermann geometry, the outer point of the vehicle front will run on the maximum radius
R
max
, whereas a point on the inner side of the vehicle at the location of the rear axle will run
on the minimum radius R
min
. Hence, it holds
R
2
max
= (R
min
+b)
2
+ (a +f)
2
, (7.5)
where a, b are the wheel base and the width of the vehicle, and f species the distance from the
front of the vehicle to the front axle. Then, the space requirement R = R
max
R
min
can be
specied as a function of the cornering radius R
min
for a given vehicle dimension
R = R
max
R
min
=
_
(R
min
+b)
2
+ (a +f)
2
R
min
. (7.6)
The space requirement R of a typical passenger car and a bus is plotted in Fig. 7.3 versus the
minimum cornering radius. In narrow curves R
min
= 5.0 m, a bus requires a space of 2.5 times
the width, whereas a passenger car needs only 1.5 times the width.
109
7 Lateral Dynamics
M
R
min
a
b
R
m
a
x
f
Figure 7.2: Space requirement
0 10 20 30 40 50
0
1
2
3
4
5
6
7
R
min
[m]

R

[
m
]
car: a=2.50 m, b=1.60 m, f=1.00 m
bus: a=6.25 m, b=2.50 m, f=2.25 m
Figure 7.3: Space requirement of a typical passenger car and bus
110
7.1.4 Vehicle Model with Trailer
7.1.4.1 Kinematics
Fig. 7.4 shows a simple lateral dynamics model for a two-axled vehicle with a single-axled
trailer. Vehicle and trailer move on a horizontal track. The position and the orientation of the
vehicle relative to the track xed frame x
0
, y
0
, z
0
is dened by the position vector to the rear
axle center
r
02,0
=
_
_
x
y
R
_
_
(7.7)
and the rotation matrix
A
02
=
_
_
cos sin 0
sin cos 0
0 0 1
_
_
. (7.8)
Here, the tire radius R is considered to be constant, and x, y as well as the yaw angle are
generalized coordinates.
K
A
1
A
2
A
3
x
1
y 1
x 2
x
3
y2
y
3
c
a
b
x
0
y
0
Figure 7.4: Kinematic model with trailer
The position vector
r
01,0
= r
02,0
+ A
02
r
21,2
with r
21,2
=
_
_
a
0
0
_
_
(7.9)
111
7 Lateral Dynamics
A
01
= A
02
A
21
with A
21
=
_
_
cos sin 0
sin cos 0
0 0 1
_
_
(7.10)
describe the position and the orientation of the front axle, where a = const labels the wheel
base and the steering angle.
The position vector
r
03,0
= r
02,0
+A
02
_
r
2K,2
+A
23
r
K3,3
_
(7.11)
with
r
2K,2
=
_
_
b
0
0
_
_
and r
K3,2
=
_
_
c
0
0
_
_
(7.12)
A
03
= A
02
A
23
with A
23
=
_
_
cos sin 0
sin cos 0
0 0 1
_
_
(7.13)
dene the position and the orientation of the trailer axis, with labeling the bend angle between
vehicle and trailer, and b, c marking the distances from the rear axle 2 to the coupling point K
and from the coupling point K to the trailer axis 3.
7.1.4.2 Vehicle Motion
According to the kinematic tire model, cf. section 7.1.1, the velocity at the rear axle can only
have a component in the longitudinal direction of the tire which here corresponds with the
longitudinal direction of the vehicle
v
02,2
=
_
_
v
x2
0
0
_
_
. (7.14)
The time derivative of Eq. (7.7) results in
v
02,0
= r
02,0
=
_
_
x
y
0
_
_
. (7.15)
The transformation of Eq. (7.14) into the system 0
v
02,0
= A
02
v
02,2
= A
02
_
_
v
x2
0
0
_
_
=
_
_
cos v
x2
sin v
x2
0
_
_
(7.16)
112
compared to Eq. (7.15) results in two rst order differential equations for the position coordi-
nates x and y
x = v
x2
cos , y = v
x2
sin . (7.17)
The velocity at the front axle follows from Eq. (7.9)
v
01,0
= r
01,0
= r
02,0
+
02,0
A
02
r
21,2
. (7.18)
Transformed into the vehicle xed system x
2
, y
2
, z
2
we obtain
v
01,2
=
_
_
v
x2
0
0
_
_
. .
v
02,2
+
_
_
0
0

_
_
. .
02,2
_
_
a
0
0
_
_
. .
r
21,2
=
_
_
v
x2
a
0
_
_
. (7.19)
The unit vectors
e
x1,2
=
_
_
cos
sin
0
_
_
and e
y1,2
=
_
_
sin
cos
0
_
_
(7.20)
dene the longitudinal and lateral direction at the front axle. According to Eq. (7.1) the velocity
component lateral to the wheel must vanish,
e
T
y1,2
v
01,2
= sin v
x2
+ cos a = 0 . (7.21)
Whereas in longitudinal direction the velocity
e
T
x1,2
v
01,2
= cos v
x2
+ sin a = v
x1
(7.22)
remains. From Eq. (7.21) a rst order differential equation follows for the yaw angle
=
v
x2
a
tan .
(7.23)
7.1.4.3 Entering a Curve
In analogy to Eq. (7.2) the steering angle can be related to the current track radius R or with
k = 1/R to the current track curvature
tan =
a
R
= a k . (7.24)
Then, the differential equation for the yaw angle reads as
= v
x2
k . (7.25)
With the curvature gradient
k = k(t) = k
C
t
T
, (7.26)
113
7 Lateral Dynamics
the entering of a curve is described as a continuous transition from a straight line with the
curvature k = 0 into a circle with the curvature k = k
C
.
The yaw angle of the vehicle can be calculated by simple integration now
(t) =
v
x2
k
C
T
t
2
2
, (7.27)
where at time t = 0 a vanishing yaw angle, (t =0) = 0, has been assumed. Then, the position
of the vehicle follows with Eq. (7.27) from the differential equations Eq. (7.17)
x = v
x2
t=T
_
t=0
cos
_
v
x2
k
C
T
t
2
2
_
dt , y = v
x2
t=T
_
t=0
sin
_
v
x2
k
C
T
t
2
2
_
dt . (7.28)
At constant vehicle speed, v
x2
= const., Eq. (7.28) is the parameterized form of a clothoide.
From Eq. (7.24) the necessary steering angle can be calculated, too. If only small steering an-
gles are necessary for driving through the curve, the tan-function can be approximated by its
argument, and
= (t) a k = a k
C
t
T
(7.29)
holds, i.e. the driving through a clothoide is manageable by a continuous steer motion.
7.1.4.4 Trailer Motions
The velocity of the trailer axis can be obtained by differentiation of the position vector Eq. (7.11)
v
03,0
= r
03,0
= r
02,0
+
02,0
A
02
r
23,2
+ A
02
r
23,2
. (7.30)
The velocity r
02,0
= v
02,0
and the angular velocity
02,0
of the vehicle are dened in Eqs. (7.16)
and (7.19). The position vector from the rear axle to the axle of the trailer is given by
r
23,2
= r
2K,2
+ A
23
r
K3,3
=
_
_
b c cos
c sin
0
_
_
, (7.31)
where r
2K,2
and r
K3,3
are dened in Eq. (7.12). The time derivative of Eq. (7.31) results in
r
23,2
=
_
_
0
0

_
_
. .
23,2
_
_
c cos
c sin
0
_
_
. .
A
23
r
K3,3
=
_
_
c sin
c cos
0
_
_
. (7.32)
Eq. (7.30) is transformed into the vehicle xed frame x
2
, y
2
, z
2
now
v
03,2
=
_
_
v
x2
0
0
_
_
. .
v
02,2
+
_
_
0
0

_
_
. .
02,2
_
_
b c cos
c sin
0
_
_
. .
r
23,2
+
_
_
c sin
c cos
0
_
_
. .
r
23,2
=
_
_
v
x2
+c sin ( + )
b c cos ( + )
0
_
_
.
(7.33)
114
The longitudinal and lateral direction at the trailer axle are dened by the unit vectors
e
x3,2
=
_
_
cos
sin
0
_
_
and e
y3,2
=
_
_
sin
cos
0
_
_
. (7.34)
At the trailer axis the lateral velocity must also vanish
e
T
y3,2
v
03,2
= sin
_
v
x2
+c sin ( + )
_
+ cos
_
b c cos ( + )
_
= 0 , (7.35)
whereas in longitudinal direction the velocity
e
T
x3,2
v
03,2
= cos
_
v
x2
+c sin ( + )
_
+ sin
_
b c cos ( + )
_
= v
x3
(7.36)
remains. If Eq. (7.23) is inserted into Eq. (7.35) now, one will get a rst order differential
equation for the bend angle
=
v
x2
a
_
a
c
sin +
_
b
c
cos + 1
_
tan
_
. (7.37)
The differential equations Eq. (7.17) and Eq. (7.23) describe position and orientation within the
x
0
, y
0
plane. The position of the trailer relative to the vehicle follows from Eq. (7.37).
7.1.4.5 Course Calculations
0 5 10 15 20 25 30
0
10
20
30
[s]
front axle steering angle
-30 -20 -10 0 10 20 30 40 50 60
0
10
20
[m]
[m]
front axle
rear axle
trailer axle
[
o
]
Figure 7.5: Entering a curve
For a given set of vehicle parameters a, b, c, and predened time functions of the vehicle ve-
locity, v
x2
= v
x2
(t) and the steering angle, = (t), the course of vehicle and trailer can be
calculated by numerical integration of the differential equations Eqs. (7.17), (7.23) and (7.37).
If the steering angle is slowly increased at constant driving speed, the vehicle drives a gure
which will be similar to a clothoide, Fig. 7.5.
115
7 Lateral Dynamics
7.2 Steady State Cornering
7.2.1 Cornering Resistance
In a body xed reference frame B, Fig. 7.6, the velocity state of the vehicle can be described by
v
0C,B
=
_
_
v cos
v sin
0
_
_
and
0F,B
=
_
_
0
0
_
_
, (7.38)
where denotes the side slip angle of the vehicle measured at the center of gravity. The angular
velocity of a vehicle cornering with constant velocity v on an at horizontal road is given by
=
v
R
, (7.39)
where R denotes the radius of curvature.
C
v
y
B
x
B
F
x1 F
y1
F
x2
F
y2
a
1
a
2
R
Figure 7.6: Cornering resistance
In the body xed reference frame, linear and angular momentum result in
m
_
v
2
R
sin
_
= F
x1
cos F
y1
sin +F
x2
, (7.40)
m
_
v
2
R
cos
_
= F
x1
sin +F
y1
cos +F
y2
, (7.41)
0 = a
1
(F
x1
sin +F
y1
cos ) a
2
F
y2
, (7.42)
116
where m denotes the mass of the vehicle, F
x1
, F
x2
, F
y1
, F
y2
are the resulting forces in longitu-
dinal and vertical direction applied at the front and rear axle, and species the average steer
angle at the front axle.
The engine torque is distributed by the center differential to the front and rear axle. Then, in
steady state condition we obtain
F
x1
= k F
D
and F
x2
= (1 k) F
D
, (7.43)
where F
D
is the driving force and by k different driving conditions can be modeled:
k = 0 rear wheel drive F
x1
= 0, F
x2
= F
D
0 < k < 1 all wheel drive
F
x1
F
x2
=
k
1 k
k = 1 front wheel drive F
x1
= F
D
, F
x2
= 0
If we insert Eq. (7.43) into Eq. (7.40) we will get
_
k cos + (1k)
_
F
D
sin F
y1
=
mv
2
R
sin ,
k sin F
D
+ cos F
y1
+ F
y2
=
mv
2
R
cos ,
a
1
k sin F
D
+ a
1
cos F
y1
a
2
F
y2
= 0 .
(7.44)
These equations can be resolved for the driving force
F
D
=
a
2
a
1
+a
2
cos sin sin cos
k + (1 k) cos
mv
2
R
. (7.45)
The driving force will vanish, if
a
2
a
1
+a
2
cos sin = sin cos or
a
2
a
1
+a
2
tan = tan (7.46)
holds. This fully corresponds with the Ackermann geometry. But, the Ackermann geometry
applies only for small lateral accelerations. In real driving situations, the side slip angle of a
vehicle at the center of gravity is always smaller than the Ackermann side slip angle. Then, due
to tan <
a
2
a
1
+a
2
tan a driving force F
D
> 0 is needed to overcome the cornering resistance
of the vehicle.
7.2.2 Overturning Limit
The overturning hazard of a vehicle is primarily determined by the track width and the height
of the center of gravity. With trucks however, also the tire deection and the body roll have to
be respected., Fig. 7.7.
117
7 Lateral Dynamics
m g
m a
y
1
2
h
2
h
1
s/2
s/2
F
zL
F
zR
F
yL
F
yR
Figure 7.7: Overturning hazard on trucks
The balance of torques at the height of the track plane applied at the already inclined vehicle
results in
(F
zL
F
zR
)
s
2
= ma
y
(h
1
+h
2
) + mg [(h
1
+h
2
)
1
+h
2
2
] , (7.47)
where a
y
describes the lateral acceleration, m is the sprung mass, and small roll angles of the
axle and the body were assumed,
1
1,
2
1.
On a left-hand tilt, the right tire raises
F
T
zR
= 0 , (7.48)
whereas the left tire carries the complete vehicle weight
F
T
zL
= mg . (7.49)
Using Eqs. (7.48) and (7.49) one gets from Eq. (7.47)
a
T
y
g
=
s
2
h
1
+h
2

T
1

h
2
h
1
+h
2
T
2
. (7.50)
The vehicle will turn over, when the lateral acceleration a
y
rises above the limit a
T
y
. Roll of axle
and body reduce the overturning limit. The angles
T
1
and
T
2
can be calculated from the tire
stiffness c
R
and the roll stiffness of the axle suspension.
118
If the vehicle drives straight ahead, the weight of the vehicle will be equally distributed to both
sides
F
stat
zR
= F
stat
zL
=
1
2
mg . (7.51)
With
F
T
zL
= F
stat
zL
+ F
z
(7.52)
and Eqs. (7.49), (7.51), one obtains for the increase of the wheel load at the overturning limit
F
z
=
1
2
mg . (7.53)
Then, the resulting tire deection follows from
F
z
= c
R
r , (7.54)
where c
R
is the radial tire stiffness.
Because the right tire simultaneously rebounds with the same amount, for the roll angle of the
axle
2 r = s
T
1
or
T
1
=
2 r
s
=
mg
s c
R
(7.55)
holds. In analogy to Eq. (7.47) the balance of torques at the body applied at the roll center of
the body yields
c
W

2
= ma
y
h
2
+ mg h
2
(
1
+
2
) , (7.56)
where c
W
names the roll stiffness of the body suspension. In particular, at the overturning limit
a
y
= a
T
y
T
2
=
a
T
y
g
mgh
2
c
W
mgh
2
+
mgh
2
c
W
mgh
2
T
1
(7.57)
applies. Not allowing the vehicle to overturn already at a
T
y
= 0 demands a minimum of roll
stiffness c
W
> c
min
W
= mgh
2
. With Eqs. (7.55) and (7.57) the overturning condition Eq. (7.50)
reads as
(h
1
+h
2
)
a
T
y
g
=
s
2
(h
1
+h
2
)
1
c
R
h
2
a
T
y
g
1
c
W
1
h
2
1
c
W
1
1
c
R
, (7.58)
where, for abbreviation purposes, the dimensionless stiffnesses
c
R
=
c
R
mg
s
and c
W
=
c
W
mg h
2
(7.59)
have been used. Resolved for the normalized lateral acceleration
a
T
y
g
=
s
2
h
1
+h
2
+
h
2
c
W
1
1
c
R
(7.60)
119
7 Lateral Dynamics
0 10 20
0
0.1
0.2
0.3
0.4
0.5
0.6
normalized roll stiffness c
W
*
0 10 20
0
5
10
15
20
T
T
normalized roll stiffness c
W
*
overturning limit a
y
roll angle =
1
+
2
Figure 7.8: Tilting limit for a typical truck at steady state cornering
remains.
At heavy trucks, a twin tire axle may be loaded with m = 13 000 kg. The radial stiffness of one
tire is c
R
= 800 000N/m, and the track width can be set to s = 2m. The values h
1
= 0.8mand
h
2
= 1.0mhold at maximal load. These values produce the results shown in Fig. 7.8. Even with
a rigid body suspension c
W
, the vehicle turns over at a lateral acceleration of a
y
0.5 g.
Then, the roll angle of the vehicle solely results from the tire deection.
At a normalized roll stiffness of c
W
= 5, the overturning limit lies at a
y
0.45g and so reaches
already 90% of the maximum. The vehicle will turn over at a roll angle of =
1
+
2
10
then.
7.2.3 Roll Support and Camber Compensation
When a vehicle drives through a curve with the lateral acceleration a
y
, centrifugal forces will
be applied to the single masses. At the simple roll model in Fig. 7.9, these are the forces m
A
a
y
and m
R
a
y
, where m
A
names the body mass and m
R
the wheel mass.
Through the centrifugal force m
A
a
y
applied to the body at the center of gravity, a torque is
generated, which rolls the body with the angle
A
and leads to an opposite deection of the
tires z
1
= z
2
.
At steady state cornering, the vehicle forces are balanced. With the principle of virtual work
W = 0 , (7.61)
the equilibrium position can be calculated.
At the simple vehicle model in Fig. 7.9 the suspension forces F
F1
, F
F2
and tire forces F
y1
, F
z1
,
F
y2
, F
z2
, are approximated by linear spring elements with the constants c
A
and c
Q
, c
R
. The work
120
F
F
1
z
1
1
y
1
F
y1
F
z1
S
1
Q
1
z
A
A
y
A
b/2 b/2
h
0
r
0
S
A
F
F
2
z
2

2
y
2
F
y2
F
y2
S
2
Q
2
m
A
a
y
m
R
a
y
m
R
a
y
Figure 7.9: Simple vehicle roll model
W of these forces can be calculated directly or using W = V via the potential V . At small
deections with linearized kinematics one gets
W = m
A
a
y
y
A
m
R
a
y
(y
A
+h
R
A
+y
1
)
2
m
R
a
y
(y
A
+h
R
A
+y
2
)
2
1
2
c
A
z
2
1

1
2
c
A
z
2
2
1
2
c
S
(z
1
z
2
)
2
1
2
c
Q
(y
A
+h
0
A
+y
1
+r
0
1
)
2
1
2
c
Q
(y
A
+h
0
A
+y
2
+r
0
2
)
2
1
2
c
R
_
z
A
+
b
2

A
+z
1
_
2
1
2
c
R
_
z
A
b
2

A
+z
2
_
2
,
(7.62)
where the abbreviation h
R
= h
0
r
0
has been used, and c
S
describes the spring constant of the
anti roll bar, converted to the vertical displacement of the wheel centers.
The kinematics of the wheel suspension are symmetrical. With the linear approaches
y
1
=
y
z
z
1
,
1
=

z

1
and y
2
=
y
z
z
2
,
2
=
z

2
(7.63)
the work W can be described as a function of the position vector
y = [ y
A
, z
A
,
A
, z
1
, z
2
]
T
. (7.64)
Due to
W = W(y) (7.65)
the principle of virtual work Eq. (7.61) leads to
W =
W
y
y = 0 . (7.66)
121
7 Lateral Dynamics
Because of y ,= 0, a system of linear equations in the form of
K y = b (7.67)
results from Eq. (7.66). The matrix K and the vector b are given by
K =
_
_
2 c
Q
0 2 c
Q
h
0
y
Q
z
c
Q

y
Q
z
c
Q
0 2 c
R
0 c
R
c
R
2 c
Q
h
0
0 c
b
2
c
R
+h
0
y
Q
z
c
Q

b
2
c
R
h
0
y
Q
z
c
Q
y
Q
z
c
Q
c
R
b
2
c
R
+h
0
y
Q
z
c
Q
c
A
+c
S
+c
R
c
S
y
Q
z
c
Q
c
R

b
2
c
R
h
0
y
Q
z
c
Q
c
S
c
A
+c
S
+c
R
_
_
(7.68)
and
b =
_
_
m
A
+ 2 m
R
0
(m
1
+m
2
) h
R
m
R
y/z
m
R
y/z
_
_
a
y
. (7.69)
The following abbreviations have been used:
y
Q
z
=
y
z
+r
0
z
, c
A
= c
A
+c
Q
_
y
z
_
2
, c
= 2 c
Q
h
2
0
+ 2 c
R
_
b
2
_
2
. (7.70)
The system of linear equations Eq. (7.67) can be solved numerically, e.g. with MATLAB. Thus,
the inuence of axle suspension and axle kinematics on the roll behavior of the vehicle can be
investigated.
A
a)
roll center
roll center
A
0
b)
0
Figure 7.10: Roll behavior at cornering: a) without and b) with camber compensation
If the wheels only move vertically to the body at jounce and rebound, at fast cornering the
wheels will be no longer perpendicular to the track Fig. 7.10 a. The camber angles
1
> 0
and
2
> 0 result in an unfavorable pressure distribution in the contact area, which leads to
a reduction of the maximally transmittable lateral forces. Thus, at more sportive vehicles axle
122
kinematics are employed, where the wheels are rotated around the longitudinal axis at jounce
and rebound,
1
=
1
(z
1
) and
2
=
2
(z
2
). Hereby, a camber compensation can be achieved
with
1
0 and
2
0. Fig. 7.10 b. By the rotation of the wheels around the longitudinal
axis on jounce and rebound, the wheel contact points are moved outwards, i.e against the lateral
force. By this, a roll support is achieved that reduces the body roll.
7.2.4 Roll Center and Roll Axis
roll center rear
roll axis
roll center front
Figure 7.11: Roll axis
The roll center can be constructed from the lateral motion of the wheel contact points Q
1
and
Q
2
, Fig. 7.10. The line through the roll center at the front and rear axle is called roll axis,
Fig. 7.11.
7.2.5 Wheel Loads
P
F0
-P
P
F0
+P
P
R0
-P
P
R0
+P
P
F0
-P
F
P
F0
+P
F
P
R0
-P
R
P
R0
+P
R
-T
T
+T
T
Figure 7.12: Wheel loads for a exible and a rigid chassis
The roll angle of a vehicle during cornering depends on the roll stiffness of the axle and on the
position of the roll center. Different axle layouts at the front and rear axle may result in different
roll angles of the front and rear part of the chassis, Fig. 7.12.
123
7 Lateral Dynamics
On most passenger cars the chassis is rather stiff. Hence, front and rear part of the chassis are
forced by an internal torque to an overall chassis roll angle. This torque affects the wheel loads
and generates different wheel load differences at the front and rear axle. Due to the degressive
inuence of the wheel load to longitudinal and lateral tire forces the steering tendency of a
vehicle can be affected.
7.3 Simple Handling Model
7.3.1 Modeling Concept
x
0
y
0
a
1
a
2
x
B
y
B
C
F
y1
F
y2
x
2
y
2
x
1
y
1
v
Figure 7.13: Simple handling model
The main vehicle motions take place in a horizontal plane dened by the earth-xed frame 0,
Fig. 7.13. The tire forces at the wheels of one axle are combined to one resulting force. Tire
torques, rolling resistance, and aerodynamic forces and torques, applied at the vehicle, are not
taken into consideration.
7.3.2 Kinematics
The vehicle velocity at the center of gravity can be expressed easily in the body xed frame x
B
,
y
B
, z
B
v
C,B
=
_
_
v cos
v sin
0
_
_
, (7.71)
where denotes the side slip angle, and v is the magnitude of the velocity.
124
The velocity vectors and the unit vectors in longitudinal and lateral direction of the axles are
needed for the computation of the lateral slips. One gets
e
x
1
,B
=
_
_
cos
sin
0
_
_
, e
y
1
,B
=
_
_
sin
cos
0
_
_
, v
01,B
=
_
_
v cos
v sin +a
1

0
_
_
(7.72)
and
e
x
2
,B
=
_
_
1
0
0
_
_
, e
y
2
,B
=
_
_
0
1
0
_
_
, v
02,B
=
_
_
v cos
v sin a
2

0
_
_
, (7.73)
where a
1
and a
2
are the distances from the center of gravity to the front and rear axle, and
denotes the yaw angular velocity of the vehicle.
7.3.3 Tire Forces
Unlike with the kinematic tire model, now small lateral motions in the contact points are per-
mitted. At small lateral slips, the lateral force can be approximated by a linear approach
F
y
= c
S
s
y
, (7.74)
where c
S
is a constant depending on the wheel load F
z
, and the lateral slip s
y
is dened by
Eq. (3.61). Because the vehicle is neither accelerated nor decelerated, the rolling condition is
fullled at each wheel
r
D
= e
T
x
v
0P
. (7.75)
Here, r
D
is the dynamic tire radius, v
0P
the contact point velocity, and e
x
the unit vector in
longitudinal direction. With the lateral tire velocity
v
y
= e
T
y
v
0P
(7.76)
and the rolling condition Eq. (7.75), the lateral slip can be calculated from
s
y
=
e
T
y
v
0P
[ e
T
x
v
0P
[
, (7.77)
with e
y
labeling the unit vector in the lateral direction direction of the tire. So, the lateral forces
are given by
F
y1
= c
S1
s
y1
; F
y2
= c
S2
s
y2
. (7.78)
7.3.4 Lateral Slips
With Eq. (7.73), the lateral slip at the front axle follows from Eq. (7.77):
s
y1
=
+sin (v cos ) cos (v sin +a
1
)
[ cos (v cos ) + sin (v sin +a
1
) [
. (7.79)
125
7 Lateral Dynamics
The lateral slip at the rear axle is given by
s
y2
=
v sin a
2

[ v cos [
. (7.80)
The yaw velocity of the vehicle , the side slip angle and the steering angle are considered
to be small
[ a
1
[ [v[ ; [ a
2
[ [v[ (7.81)
[ [ 1 and [ [ 1 . (7.82)
Because the side slip angle always labels the smaller angle between the velocity vector and the
vehicle longitudinal axis, instead of v sin v the approximation
v sin [v[ (7.83)
has to be used. Now, Eqs. (7.79) and (7.80) result in
s
y1
=
a
1
[v[
+
v
[v[
(7.84)
and
s
y2
= +
a
2
[v[
, (7.85)
where the consequences of Eqs. (7.81), (7.82), and (7.83) were already taken into consideration.
7.3.5 Equations of Motion
The velocities, angular velocities, and the accelerations are needed to derive the equations of
motion, For small side slip angles 1, Eq. (7.71) can be approximated by
v
C,B
=
_
_
v
[v[
0
_
_
. (7.86)
The angular velocity is given by
0F,B
=
_
_
0
0

_
_
. (7.87)
If the vehicle accelerations are also expressed in the vehicle xed frame x
F
, y
F
, z
F
, one will
nd at constant vehicle speed v = const and with neglecting small higher-order terms
a
C,B
=
0F,B
v
C,B
+ v
C,B
=
_
_
0
v +[v[

0
_
_
. (7.88)
126
The angular acceleration is given by

0F,B
=
_
_
0
0

_
_
, (7.89)
where the substitution
= (7.90)
was used. The linear momentum in the lateral direction of the vehicle reads as
m(v +[v[

) = F
y1
+F
y2
, (7.91)
where, due to the small steering angle, the term F
y1
cos has been approximated by F
y1
, and
m describes the vehicle mass. With Eq. (7.90) the angular momentum yields
= a
1
F
y1
a
2
F
y2
, (7.92)
where names the inertia of vehicle around the vertical axis. With the linear description of the
lateral forces Eq. (7.78) and the lateral slips Eqs. (7.84), (7.85), one gets from Eqs. (7.91) and
(7.92) two coupled, but linear rst order differential equations
=
c
S1
m[v[
_

a
1
[v[
+
v
[v[
_
+
c
S2
m[v[
_
+
a
2
[v[
_

v
[v[
(7.93)
=
a
1
c
S1
_

a
1
[v[
+
v
[v[
_

a
2
c
S2
_
+
a
2
[v[
_
, (7.94)
which can be written in the form of a state equation
_

_
. .
x
=
_
c
S1
+c
S2
m[v[
a
2
c
S2
a
1
c
S1
m[v[[v[

v
[v[
a
2
c
S2
a
1
c
S1

a
2
1
c
S1
+a
2
2
c
S2
[v[
_
_
. .
A
_

_
. .
x
+
_
_
v
[v[
c
S1
m[v[
v
[v[
a
1
c
S1
_
. .
B
_
..
u
. (7.95)
If a system can be at least approximatively described by a linear state equation, stability, steady
state solutions, transient response, and optimal controlling can be calculated with classic meth-
ods of system dynamics.
7.3.6 Stability
7.3.6.1 Eigenvalues
The homogeneous state equation
x = Ax (7.96)
127
7 Lateral Dynamics
describes the eigen-dynamics. If the approach
x
h
(t) = x
0
e
t
(7.97)
is inserted into Eq. (7.96), the homogeneous equation will remain
(E A) x
0
= 0 . (7.98)
One gets non-trivial solutions x
0
,= 0 for
det [E A[ = 0 . (7.99)
The eigenvalues provide information concerning the stability of the system.
7.3.6.2 Low Speed Approximation
The state matrix
A
v0
=
_
c
S1
+c
S2
m[v[
a
2
c
S2
a
1
c
S1
m[v[[v[

v
[v[
0
a
2
1
c
S1
+a
2
2
c
S2
[v[
_
_
(7.100)
approximates the eigen-dynamics of vehicles at low speeds, v 0. The matrix in Eq. (7.100)
has the eigenvalues
1
v0
=
c
S1
+c
S2
m[v[
and
2
v0
=
a
2
1
c
S1
+a
2
2
c
S2
[v[
. (7.101)
The eigenvalues are real and always negative independent from the driving direction. Thus,
vehicles possess an asymptotically stable driving behavior at low speed!
7.3.6.3 High Speed Approximation
At high driving velocities, v , the state matrix can be approximated by
A
v
=
_
_
0
v
[v[
a
2
c
S2
a
1
c
S1
0
_
_
. (7.102)
Using Eq. (7.102) one receives from Eq. (7.99) the relation
2
v
+
v
[v[
a
2
c
S2
a
1
c
S1
= 0 (7.103)
128
with the solutions
1,2
v
=
_
v
[v[
a
2
c
S2
a
1
c
S1
. (7.104)
When driving forward with v > 0, the root argument will be positive, if
a
2
c
S2
a
1
c
S1
< 0 (7.105)
holds. Then however, one eigenvalue is positive, and the system is unstable. Two zero-
eigenvalues
1
= 0 and
2
= 0 are obtained for
a
1
c
S1
= a
2
c
S2
. (7.106)
The driving behavior is indifferent then. Slight parameter variations, however, can lead to an
unstable behavior. With
a
2
c
S2
a
1
c
S1
> 0 or a
1
c
S1
< a
2
c
S2
(7.107)
and v > 0 the root argument in Eq. (7.104) becomes negative. Then, the eigenvalues are imag-
inary, and disturbances lead to undamped vibrations. To avoid instability, high-speed vehicles
have to satisfy the condition Eq. (7.107). The root argument in Eq. (7.104) changes at backward
driving its sign. Hence, a vehicle showing stable driving behavior at forward driving becomes
unstable at fast backward driving!
7.3.6.4 Critical Speed
The condition for non-trivial solutions (7.99) results here in a quadratic equation for the eigen-
values
det [E A[ =
2
+k
1
+k
2
= 0 (7.108)
which is solved by
1,2
=
k
1
2

_
k
1
2
_
2
k
2
. (7.109)
Hence, asymptotically stable solutions demand for
k
1
> 0 and k
2
> 0 (7.110)
which corresponds with the stability criteria of Stodola and Hurwitz [23].
According to Eq. (7.95) the coefcients in Eq. (7.108) can be derived from the vehicle data
k
1
=
c
S1
+c
S2
m[v[
+
a
2
1
c
S1
+a
2
2
c
S2
[v[
, (7.111)
k
2
=
c
S1
+c
S2
m[v[
a
2
1
c
S1
+a
2
2
c
S2
[v[

(a
2
c
S2
a
1
c
S1
)
2
m[v[[v[
+
v
[v[
a
2
c
S2
a
1
c
S1
=
c
S1
c
S2
(a
1
+a
2
)
2
mv
2
_
1 +
v
[v[
a
2
c
S2
a
1
c
S1
c
S1
c
S2
(a
1
+a
2
)
2
mv
2
_
.
(7.112)
129
7 Lateral Dynamics
The coefcient k
1
is always positive, whereas k
2
> 0 is fullled only if
1 +
v
[v[
a
2
c
S2
a
1
c
S1
c
S1
c
S2
(a
1
+a
2
)
2
mv
2
> 0 (7.113)
will hold. Hence, a vehicle designed stable for arbitrary velocities in forward direction becomes
unstable, when it drives too fast backwards. Because, k
2
> 0 for a
2
c
S2
a
1
c
S1
> 0 and v < 0
demands for v > v
C
, where according to Eq. (7.113) the critical backwards velocity is given
by
v
C
=
c
S1
c
S2
(a
1
+a
2
)
2
m(a
2
c
S2
a
1
c
S1
)
. (7.114)
On the other hand, vehicle layouts with a
2
c
S2
a
1
c
S1
< 0 or are only stable while driving
forward as long as v < v
+
C
will hold. Here, Eq. (7.113) yields the critical forward velocity of
v
+
C
=
c
S1
c
S2
(a
1
+a
2
)
2
m(a
1
c
S1
a
2
c
S2
)
. (7.115)
Most vehicles are designed stable for fast forward drive. Then, the backwards velocity must be
limited in order to avoid stability problems. That is why, fast driving vehicles have four or more
gears for forward drive but, only one or two reverse gears.
7.3.7 Steady State Solution
7.3.7.1 Steering Tendency
At a given steering angle =
0
, a stable system reaches steady state after a certain time. Then,
the vehicle will drive on a circle with the radius R
st
which is determined by
st
=
v
R
st
(7.116)
where v is the velocity of the vehicle and
st
denotes its steady state angular velocity.
With x
st
= const. or x
st
= 0, the state equation Eq. (7.95) is reduced to a system of linear
equations
Ax
st
= Bu . (7.117)
Using Eq. (7.116) the state vector can be described in steady state by
x
st
=
_

st
v/R
st
_
, (7.118)
where
st
denotes the steady state side slip angle. With u = [
0
], and the elements of the state
matrix Aand the vector B which are dened in Eq. (7.95) the system of linear equations (7.117)
yields
(c
S1
+c
S2
)
st
+ (mv [v[ +a
1
c
S1
a
2
c
S2
)
v
[v[
1
R
st
=
v
[v[
c
S1
0
, (7.119)
130
(a
1
c
S1
a
2
c
S2
)
st
+ (a
2
1
c
S1
+a
2
2
c
S2
)
v
[v[
1
R
st
=
v
[v[
a
1
c
S1
0
, (7.120)
where the rst equation has been multiplied by m[v[ and the second with . Eliminating
the steady state side slip angle
st
leads to
_
mv[v[(a
1
c
S1
a
2
c
S2
) + (a
1
c
S1
a
2
c
S2
)
2
(c
S1
+c
S2
)(a
2
1
c
S1
+a
2
2
c
S2
)
v
[v[
1
R
st
=
[a
1
c
S1
a
2
c
S2
a
1
(c
S1
+c
S2
)]
v
[v[
c
S1
0
,
(7.121)
which can be simplied to
_
mv[v[(a
1
c
S1
a
2
c
S2
) c
S1
c
S2
(a
1
+a
2
)
2
v
[v[
1
R
st
=
v
[v[
c
S1
c
S2
(a
1
+a
2
)
0
. (7.122)
Hence, driving the vehicle at a certain radius requires a steering angle of
0
=
a
1
+a
2
R
st
+ m
v[v[
R
st
a
2
c
S2
a
1
c
S1
c
S1
c
S2
(a
1
+a
2
)
. (7.123)
The rst term is the Ackermann steering angle which follows from Eq. (7.2) with the wheel
base a = a
1
+a
2
and the approximation for small steering angles tan
0
0
. The Ackermann-
steering angle provides a good approximation for slowly moving vehicles, because the second
expression in Eq. (7.123) becomes very small at v 0. Depending on the value of a
2
c
S2

a
1
c
S1
and the driving direction, forward v > 0 or backward v < 0, the necessary steering angle
differs from the Ackermann-steering angle at higher speeds. The difference is proportional to
the lateral acceleration
a
y
=
v[v[
R
st
=
v
2
R
st
. (7.124)
Hence, Eq. (7.123) can be written as
0
=
A
+ k
v
2
R
st
, (7.125)
where
A
=
a
1
+a
2
R
st
is the Ackermann steering angle, and k summarizes the relevant vehicle
parameter. In a diagram where the steering angle
0
is plotted versus the lateral acceleration
a
y
= v
2
/R
st
Eq. (7.125) represents a straight line , Fig. 7.14.
On forward drive, v > 0, the inclination of the line is given by
k =
m (a
2
c
S2
a
1
c
S1
)
c
S1
c
S2
(a
1
+a
2
)
. (7.126)
At steady state cornering the amount of the steering angle
0
<
=
>

A
and hence, the steering
tendency depends at increasing velocity on the stability condition a
2
c
S2
a
1
c
S1
<
=
>
0. The
various steering tendencies are also arranged in Tab. 7.1.
131
7 Lateral Dynamics
a
y
= v
2
/R
st
A
oversteering:
0
<
A
or a
1
c
S1
> a
2
c
S2
0
neutral:
0
=
A
or a
1
c
S1
= a
2
c
S2
understeering:
0
>
A
or a
1
c
S1
< a
2
c
S2
Figure 7.14: Steering angle versus lateral acceleration
understeering
0
>
A
0
or a
1
c
S1
< a
2
c
S2
or a
1
c
S1
/ a
2
c
S2
< 1
neutral
0
=
A
0
or a
1
c
S1
= a
2
c
S2
or a
1
c
S1
/ a
2
c
S2
= 1
oversteering
0
<
A
0
or a
1
c
S1
> a
2
c
S2
or a
1
c
S1
/ a
2
c
S2
> 1
Table 7.1: Steering tendencies of a vehicle at forward driving
7.3.7.2 Side Slip Angle
Equations (7.119) and (7.120) can also be resolved for the steady state side slip angle. One gets
st
=
v
[v[
a
2
mv [v[
a
1
c
S2
(a
1
+a
2
)
a
1
+a
2
+ mv [v[
a
2
c
S2
a
1
c
S1
c
S1
c
S2
(a
1
+a
2
)
0
, (7.127)
The steady state side slip angle starts with the kinematic value
v0
st
=
v
[v[
a
2
a
1
+a
2
0
. (7.128)
On forward drive v > 0 it decreases with increasing speed till the side slip angle changes the
sign at
v
st
=0
=
a
2
c
S2
(a
1
+a
2
)
a
1
m
. (7.129)
In Fig. 7.15 the side slip angle , and the driven curve radius R are plotted versus the driving
speed v. The steering angle has been set to
0
= 1.4321
, in order to let the vehicle drive a circle

with the radius R
0
= 100 m at v 0. The actually driven circle radius r = R
st
(
0
) has been
calculated from Eq. (7.123).
132
0 10 20 30 40
-10
-8
-6
-4
-2
0
2
v [m/s]

[
d
e
g
]
steady state side slip angle
a
1
*
c
S1
/a
2
*
c
S2
= 0.66667
a
1
*
c
S1
/a
2
*
c
S2
= 1
a
1
*
c
S1
/a
2
*
c
S2
= 1.3333
0 10 20 30 40
0
50
100
150
200
v [m/s]
r

[
m
]
radius of curvrature
a
1
*
c
S1
/a
2
*
c
S2
= 0.66667
a
1
*
c
S1
/a
2
*
c
S2
= 1
a
1
*
c
S1
/a
2
*
c
S2
= 1.3333
m=700 kg;
=1000 kg m
2
;
a
1
=1.2 m;
a
2
=1.3 m;
c
S1
= 80 000 Nm; c
S2
=
110 770 Nm
73 846 Nm
55 385 Nm
Figure 7.15: Side slip angle at steady state cornering
Some concepts for an additional steering of the rear axle were trying to keep the side slip
angle of the vehicle, measured at the center of the vehicle to zero by an appropriate steering or
controlling. Due to numerous problems, production stage could not yet be reached.
7.3.7.3 Slip Angles
With the conditions for a steady state solution

st
= 0,
st
= 0 and Eq. (7.116), the equations
of motion Eq. (7.91) and Eq. (7.92) can be resolved for the lateral forces
F
y1
st
=
a
2
a
1
+a
2
m
v
2
R
st
,
F
y2
st
=
a
1
a
1
+a
2
m
v
2
R
st
or
a
1
a
2
=
F
y2
st
F
y1
st
. (7.130)
With the linear tire model in Eq. (7.74) one gets in addition
F
st
y1
= c
S1
s
st
y1
and F
st
y2
= c
S2
s
st
y2
, (7.131)
where s
st
y
A1
and s
st
y
A2
label the steady state lateral slips at the front and rear axle. Now, from
Eqs. (7.130) and (7.131) it follows
a
1
a
2
=
F
st
y2
F
st
y1
=
c
S2
s
st
y2
c
S1
s
st
y1
or
a
1
c
S1
a
2
c
S2
=
s
st
y2
s
st
y1
. (7.132)
That means, at a vehicle with a tendency to understeer (a
1
c
S1
< a
2
c
S2
) during steady state
cornering the slip angles at the front axle are larger than the slip angles at the rear axle, s
st
y1
>
s
st
y2
. So, the steering tendency can also be determined from the slip angle at the axles.
133
7 Lateral Dynamics
7.3.8 Inuence of Wheel Load on Cornering Stiffness
With identical tires at the front and rear axle, given a linear inuence of wheel load on the raise
of the lateral force over the lateral slip,
c
lin
S1
= c
S
F
z1
and c
lin
S2
= c
S
F
z2
. (7.133)
holds. The weight of the vehicle G = mg is distributed over the axles according to the position
of the center of gravity
F
z1
=
a
2
a
1
+a
2
G and .F
z2
=
a
1
a
1
+a
2
G (7.134)
With Eq. (7.133) and Eq. (7.134) one obtains
a
1
c
lin
S1
= a
1
c
S
a
2
a
1
+a
2
G (7.135)
and
a
2
c
lin
S2
= a
2
c
S
a
1
a
1
+a
2
G . (7.136)
Thus, a vehicle with identical tires would be steering neutrally at a linear inuence of the wheel
load on the cornering stiffness, because of
a
1
c
lin
S1
= a
2
c
lin
S2
(7.137)
The lateral force is applied behind the center of the contact patch at the caster offset distance.
Hence, the lever arms of the lateral forces change to a
1
a
1
v
|v|
n
L
1
and a
2
a
2
+
v
|v|
n
L
1
,
which will stabilize the vehicle, independently from the driving direction.
0 1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
F
z
[kN]
F
y

[
k
N
]
F
z
[N] F
y
[N]
0 0
1000 758
2000 1438
3000 2043
4000 2576
5000 3039
6000 3434
7000 3762
8000 4025
Figure 7.16: Lateral force F
y
over wheel load F
z
at different slip angles
At a real tire, a degressive inuence of the wheel load on the tire forces is observed, Fig. 7.16.
According to Eq. (7.92) the rotation of the vehicle is stable, if the torque from the lateral forces
F
y1
and F
y2
is aligning, i.e.
a
1
F
y1
a
2
F
y2
< 0 (7.138)
134
holds. At a vehicle with the wheel base a = 2.45 m the axle loads F
z1
= 4000 N and F
z2
=
3000 N yield the position of the center of gravity a
1
= 1.05 m and a
2
= 1.40 m. At equal slip
on front and rear axle one gets from the table in 7.16 F
y1
= 2576 N and F
y2
= 2043 N. With
this, the condition in Eq. (7.138) yields 1.05 2576 1.45 2043 = 257.55 . The value is
signicantly negative and thus stabilizing.
Vehicles with a
1
< a
2
have a stable, i.e. understeering driving behavior. If the axle load at the
rear axle is larger than at the front axle (a
1
> a
2
), generally a stable driving behavior can only
be achieved with different tires.
At increasing lateral acceleration the vehicle is more and more supported by the outer wheels.
The wheel load differences can differ at a sufciently rigid vehicle body, because of different
kinematics (roll support) or different roll stiffness. Due to the degressive inuence of wheel
load, the lateral force at an axle decreases with increasing wheel load difference. If the wheel
load is split more strongly at the front axle than at the rear axle, the lateral force potential at the
front axle will decrease more than at the rear axle and the vehicle will become more stable with
an increasing lateral force, i.e. more understeering.
135
8 Driving Behavior of Single Vehicles
8.1 Standard Driving Maneuvers
8.1.1 Steady State Cornering
The steering tendency of a real vehicle is determined by the driving maneuver called steady
state cornering. The maneuver is performed quasi-static. The driver tries to keep the vehicle on
a circle with the given radius R. He slowly increases the driving speed v and, with this also
the lateral acceleration due a
y
=
v
2
R
until reaching the limit. Typical results are displayed in
Fig. 8.1.
0
20
40
60
80
lateral acceleration [g]
s
t
e
e
r

a
n
g
l
e

[
d
e
g
]
-4
-2
0
2
4
s
i
d
e

s
l
i
p

a
n
g
l
e

[
d
e
g
]
0 0.2 0.4 0.6 0.8
0
1
2
3
4
r
o
l
l

a
n
g
l
e

[
d
e
g
]
0 0.2 0.4 0.6 0.8
0
1
2
3
4
5
6
w
h
e
e
l

l
o
a
d
s

[
k
N
]
lateral acceleration [g]
Figure 8.1: Steady state cornering: rear-wheel-driven car on R = 100 m
In forward drive the vehicle is understeering and thus stable for any velocity. The inclination
in the diagram steering angle versus lateral velocity decides about the steering tendency and
stability behavior.
136
The nonlinear inuence of the wheel load on the tire performance is here used to design a vehicle
that is weakly stable, but sensitive to steer input in the lower range of lateral acceleration, and
is very stable but less sensitive to steer input in limit conditions.
With the increase of the lateral acceleration the roll angle becomes larger. The overturning
torque is intercepted by according wheel load differences between the outer and inner wheels.
With a sufciently rigid frame the use of an anti roll bar at the front axle allows to increase the
wheel load difference there and to decrease it at the rear axle accordingly.
Thus, the digressive inuence of the wheel load on the tire properties, cornering stiffness and
maximum possible lateral force, is stressed more strongly at the front axle, and the vehicle
becomes more under-steering and stable at increasing lateral acceleration, until it drifts out of
the curve over the front axle in the limit situation.
Problems occur at front driven vehicles, because due to the demand for traction, the front axle
cannot be relieved at will.
Having a sufciently large test site, the steady state cornering maneuver can also be carried out
at constant speed. There, the steering wheel is slowly turned until the vehicle reaches the limit
range. That way also weakly motorized vehicles can be tested at high lateral accelerations.
8.1.2 Step Steer Input
The dynamic response of a vehicle is often tested with a step steer input. Methods for the
calculation and evaluation of an ideal response, as used in system theory or control technics,
can not be used with a real car, for a step input at the steering wheel is not possible in practice.
A real steering angle gradient is displayed in Fig. 8.2.
0 0.2 0.4 0.6 0.8 1
0
10
20
30
40
time [s]
s
t
e
e
r
i
n
g

a
n
g
l
e

[
d
e
g
]
Figure 8.2: Step Steer Input
Not the angle at the steering wheel is the decisive factor for the driving behavior, but the steering
angle at the wheels, which can differ from the steering wheel angle because of elasticities,
friction inuences, and a servo-support. At very fast steering movements, also the dynamics of
the tire forces plays an important role.
In practice, a step steer input is usually only used to judge vehicles subjectively. Exceeds in yaw
velocity, roll angle, and especially sideslip angle are felt as annoying.
137
0
0.1
0.2
0.3
0.4
0.5
0.6
l
a
t
e
r
a
l

a
c
c
e
l
e
r
a
t
i
o
n

[
g
]
0
2
4
6
8
10
12
y
a
w

v
e
l
o
c
i
t
y

[
d
e
g
/
s
]
0 2 4
0
0.5
1
1.5
2
2.5
3
r
o
l
l

a
n
g
l
e

[
d
e
g
]
0 2 4
-2
-1.5
-1
-0.5
0
0.5
1
[t]
s
i
d
e

s
l
i
p

a
n
g
l
e

[
d
e
g
]
Figure 8.3: Step Steer: Passenger Car at v = 100 km/h
The vehicle under consideration behaves dynamically very well, Fig. 8.3. Almost no overshoots
occur in the time history of the roll angle and the lateral acceleration. However, small overshoots
can be noticed at yaw the velocity and the sideslip angle.
8.1.3 Driving Straight Ahead
8.1.3.1 Random Road Prole
The irregularities of a track are of stochastic nature. Fig. 8.4 shows a country road prole in
different scalings. To limit the effort of the stochastic description of a track, one usually employs
simplifying models. Instead of a fully two-dimensional description either two parallel tracks are
evaluated
z = z(x, y) z
1
= z
1
(s
1
) , and z
2
= z
2
(s
2
) (8.1)
or one uses an isotropic track. The statistic properties are direction-independent at an isotropic
track. Then, a two-dimensional track can be approximated by a single random process
z = z(x, y) z = z(s) ; (8.2)
138
0 10 20 30 40 50 60 70 80 90 100
0
1
2
3
4
5
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
0.04
0.05
Figure 8.4: Track Irregularities
A normally distributed, stationary and ergodic random process z = z(s) is completely charac-
terized by the rst two expectation values, the mean value
m
z
= lim
s
1
2s
s
_
s
z(s) ds (8.3)
and the correlation function
R
zz
() = lim
s
1
2s
s
_
s
z(s) z(s ) ds . (8.4)
A vanishing mean value m
z
= 0 can always be achieved by an appropriate coordinate transfor-
mation. The correlation function is symmetric,
R
zz
() = R
zz
() , (8.5)
and
R
zz
(0) = lim
s
1
2s
s
_
s
_
z(s)
_
2
ds (8.6)
describes the variance of z
s
.
Stochastic track irregularities are mostly described by power spectral densities (abbreviated by
psd). Correlating function and the one-sided power spectral density are linked by the Fourier-
transformation
R
zz
() =
_
0
S
zz
() cos() d (8.7)
where denotes the space circular frequency. With Eq. (8.7) follows from Eq. (8.6)
R
zz
(0) =
_
0
S
zz
() d . (8.8)
139
Thus, the psd gives information, how the variance is compiled from the single frequency shares.
The power spectral densities of real tracks can be approximated by the relation
S
zz
() = S
0
_

0
_
w
, (8.9)
where the reference frequency is xed to
0
= 1 m
1
. The reference psd S
0
= S
zz
(
0
) acts
as a measurement for unevennes and the waviness w indicates, whether the track has notable
irregularities in the short or long wave spectrum. At real tracks, the reference-psd S
0
lies within
the range from 1 10
6
m
3
to 100 10
6
m
3
and the waviness can be approximated by w = 2.
8.1.3.2 Steering Activity
-2 0 2
0
500
1000
highway: S
0
=1*10
-6
m
3
; w=2
-2 0 2
0
500
1000
country road: S
0
=2*10
-5
m
3
; w=2
[deg]
[deg]
Figure 8.5: Steering activity on different roads
A straightforward drive upon an uneven track makes continuous steering corrections necessary.
The histograms of the steering angle at a driving speed of v = 90km/h are displayed in Fig. 8.5.
The track quality is reected in the amount of steering actions. The steering activity is often used
to judge a vehicle in practice.
8.2 Coach with different Loading Conditions
8.2.1 Data
The difference between empty and laden is sometimes very large at trucks and coaches. In the
table 8.1 all relevant data of a travel coach in fully laden and empty condition are listed.
The coach has a wheel base of a = 6.25 m. The front axle with the track width s
v
= 2.046 m
has a double wishbone single wheel suspension. The twin-tire rear axle with the track widths
s
o
h
= 2.152 m and s
i
h
= 1.492 m is guided by two longitudinal links and an a-arm. The air-
springs are tted to load variations via a niveau-control.
140
8.2 Coach with different Loading Conditions
vehicle mass [kg] center of gravity [m] inertias [kg m
2
]
empty 12 500 3.800 [ 0.000 [ 1.500
12 500 0 0
0 155 000 0
0 0 155 000
fully laden 18 000 3.860 [ 0.000 [ 1.600
15 400 0 250
0 200 550 0
250 0 202 160
Table 8.1: Data for a laden and empty coach
-1 0 1
-10
-5
0
5
10
s
u
s
p
e
n
s
i
o
n

t
r
a
v
e
l

[
c
m
]
steer angle [deg]
Figure 8.6: Roll steer: - - front, rear
8.2.2 Roll Steering
While the kinematics at the front axle hardly cause steering movements at roll motions, the
kinematics at the rear axle are tuned in a way to cause a notable roll steering effect, Fig. 8.6.
8.2.3 Steady State Cornering
Fig. 8.7 shows the results of a steady state cornering on a 100 m-Radius. The fully occupied
vehicle is slightly more understeering than the empty one. The higher wheel loads cause greater
tire aligning torques and increase the degressive wheel load inuence on the increase of the
lateral forces. Additionally roll steering at the rear axle occurs.
Both vehicles can not be kept on the given radius in the limit range. Due to the high position of
the center of gravity the maximal lateral acceleration is limited by the overturning hazard. At
the empty vehicle, the inner front wheel lift off at a lateral acceleration of a
y
0.4 g . If the
vehicle is fully occupied, this effect will occur already at a
y
0.35 g.
141
0 0.1 0.2 0.3 0.4
50
100
150
200
250
lateral acceleration a
y
[g]
steer angle
LW
[deg]
-100 0 100
0
50
100
150
200
[m]
[
m
]
vehicle course
0 0.1 0.2 0.3 0.4
0
50
100
wheel loads [kN]
0 0.1 0.2 0.3 0.4
0
50
100
wheel loads [kN]
y
[g]
y
[g]
Figure 8.7: Steady State Cornering: Coach - - empty, fully occupied
0 2 4 6 8
0
0.1
0.2
0.3
0.4
y
[g]
0 2 4 6 8
0
2
4
6
8
10
yaw velocity
Z
[deg/s]
0 2 4 6 8
0
2
4
6
8
[s]
roll angle [deg]
0 2 4 6 8
-2
-1
0
1
2
[s]
side slip angle [deg]
Figure 8.8: Step steer input: - - coach empty, coach fully occupied
142
8.3 Different Rear Axle Concepts for a Passenger Car
8.2.4 Step Steer Input
The results of a step steer input at the driving speed of v = 80 km/h can be seen in Fig. 8.8. To
achieve comparable acceleration values in steady state condition, the step steer input was done
at the empty vehicle with = 90
and at the fully occupied one with = 135
. The steady state

roll angle is 50% larger at the fully occupied bus than at the empty one. By the niveau-control,
the air spring stiffness increases with the load. Because the damper effect remains unchanged,
the fully laden vehicle is not damped as well as the empty one. This results in larger overshoots
in the time histories of the lateral acceleration, the yaw angular velocity, and the sideslip angle.
8.3 Different Rear Axle Concepts for a Passenger Car
A medium-sized passenger car is equipped in standard design with a semi-trailing rear axle.
By accordingly changed data this axle can easily be transformed into a trailing arm or a single
wishbone axis. According to the roll support, the semi-trailing axle realized in serial production
represents a compromise between the trailing arm and the single wishbone, Fig. 8.9, .
-5 0 5
-10
-5
0
5
10
lateral motion [cm]
v
e
r
t
i
c
a
l

m
o
t
i
o
n

[
c
m
]
Figure 8.9: Rear axle: semi-trailing arm, - - single wishbone, trailing arm
The inuences on the driving behavior at steady state cornering on a 100 m radius are shown in
Fig. 8.10.
Substituting the semi-trailing arm at the standard car by a single wishbone, one gets, without
adaption of the other system parameters a vehicle oversteering in the limit range. Compared to
the semi-trailing arm the single wishbone causes a notably higher roll support. This increases
the wheel load difference at the rear axle, Fig. 8.10. Because the wheel load difference is simul-
taneously reduced at the front axle, the understeering tendency is reduced. In the limit range,
this even leads to an oversteering behavior.
The vehicle with a trailing arm rear axle is, compared to the serial car, more understeering. The
lack of roll support at the rear axle also causes a larger roll angle.
143
0 0.2 0.4 0.6 0.8
0
50
100
steer angle
LW
[deg]
0 0.2 0.4 0.6 0.8
0
1
2
3
4
5
roll angle [Grad]
0 0.2 0.4 0.6 0.8
0
2
4
6
wheel loads front [kN]
0 0.2 0.4 0.6 0.8
0
2
4
6
y
[g]
wheel loads rear [kN]
y
[g]
Figure 8.10: Steady state cornering, semi-trailing arm, - - single wishbone, trailing arm
144
Bibliography
[1] Bestle, D.; Befnger, M.: Design of Progressive Automotive Shock Absorbers. In: Pro-
ceedings of Multibody Dynamics 2005, Madrid 2005.
[2] Blundell, M.; Harty, D.: The Multibody System Approach to Vehicle Dynamics. Elsevier
Butterworth-Heinemann Publications, 2004.
[3] Braun, H.: Untersuchung von Fahrbahnunebenheiten und Anwendung der Ergebnisse.
Diss. TU Braunschweig 1969.
[4] Butz, T.; Ehmann, M.; Wolter, T.-M.: A Realistic Road Model for Real-Time Vehicle
Dynamics Simulation. Society of Automotive Engineers, SAE Paper 2004-01-1068, 2004.
[5] Butz, T.; von Stryk, O.; Chucholowski, C.; Truskawa,S.; Wolter,T.-M.: Modeling Tech-
niques and Parameter Estimation for the Simulation of Complex Vehicle Structures. In:
M. Breuer, F. Durst, C. Zenger (eds.): High-Performance Scientic and Engineering Com-
puting. Proceedings of the 3rd International FORTWIHR Conference, Erlangen, 12.-14.
Mrz 2001. Lecture Notes in Computational Science and Engineering 21. Springer Verlag,
2002, S. 333-340.
[6] Dodds, C. J.; Robson, J. D.;: The Description of Road Surface Roughness, J. of Sound and
Vibr. 31 (2) 1973, pp. 175-183.
[7] Dorato, P.; Abdallah, C.; Cerone, V.: Linear-Quadratic Control. An Introduction. Prentice-
Hall, Englewood Cliffs, New Jersey, 1995.
[8] Eichler, M.; Lion, A.; Sonnak, U.; Schuller, R.: Dynamik von Luftfedersystemen mit
Zusatzvolumen: Modellbildung, Fahrzeugsimulationen und Potenzial. VDI-Bericht 1791,
2003.
[9] Flexible Ring Tire Model Documentation and Users Guide. Cosin Consulting 2004,
http://www.ftire.com.
[10] Gillespie, Th.D.: Fundamentals of Vehicle Dynamics. Warrendale: Society of Automotive
Engineers, Inc., 1992.
[11] Hirschberg, W; Rill, G. Weinfurter, H.: User-Appropriate Tyre-Modeling for Vehicle Dy-
namics in Standard and Limit Situations. Vehicle System Dynamics 2002, Vol. 38, No. 2,
pp. 103-125. Lisse: Swets & Zeitlinger.
145
Bibliography
[12] Hirschberg, W., Weinfurter, H., Jung, Ch.: Ermittlung der Potenziale zur LKW-
Stabilisierung durch Fahrdynamiksimulation. VDI-Berichte 1559 Berechnung und Sim-
ulation im Fahrzeugbau Wrzburg, 14.-15. Sept. 2000.
[13] ISO 8608: Mechanical Vibration - Road Surface Proles - Reporting of Measured Data.
International Standard (ISO) 1995.
[14] van der Jagt, P.: The Road to Virtual Vehicle Prototyping; new CAE-models for accel-
erated vehicle dynamics development. PhD-Thesis, Tech. Univ. Eindhoven, Eindhoven
2000, ISBN 90-386-2552-9 NUGI 834.
[15] Kiencke, U.; Nielsen, L.: Automotive Control Systems. Berlin: Springer, 2000.
[16] Kortm, W., Lugner, P.: Systemdynamik und Regelung von Fahrzeugen. Springer Verlag,
Berlin 1993.
[17] Kosak, W.; Reichel, M.: Die neue Zentral-Lenker-Hinterachse der BMW 3er-Baureihe.
Automobiltechnische Zeitschrift, ATZ 93 (1991) 5.
[18] Lugner, P.; Pacejka, H.; Plchl,M.: Recent advances in tyre models and testing procedures.
Vehicle System Dynamics, Vol. 43, No. 67, June
UJuly 2005, 413
U436.
[19] Matschinsky, W.: Radfhrungen der Straenfahrzeuge. Berlin: Springer, 2. Au., 1998.
[20] Mitschke, M.; Wallentowitz, H.: Dynamik der Kraftfahrzeuge. 4. Auage. Springer-Verlag
Berlin Heidelberg 2004.
[21] Mller, P.C.; Popp, K.: Kovarianzanalyse von linearen Zufallsschwingungen mit zeitlich
verschobenen Erregerprozessen. Z. Angew. Math. Mech. 59 (1979), pp T144-T146.
[22] Mller, P.C.; Popp, K.; Schiehlen, W.O.: Covariance Analysis of Nonlinear Stochastic
Guideway-Vehicle-Systems. In: The Dynamics of Vehicles, Ed. Willumeit, H.P., Swets &
Zeitlinger, Lisse 1980.
[23] Mller, P.C.; Schiehlen, W.O.: Lineare Schwingungen. Wiesbaden: Akad. Verlagsge-
sellschaft 1976.
[24] Neureder, U.: Untersuchungen zur bertragung von Radlastschwankungen auf die
Lenkung von Pkw mit Federbeinvorderachse und Zahnstangenlenkung. Fortschritt-
Berichte VDI, Reihe 12, Nr. 518. Dsseldorf: VDI Verlag 2002.
[25] Oertel, Ch.; Fandre, A.: Ride Comfort Simulations an Steps Towards Life Time Calcula-
tions; RMOD-K and ADAMS. International ADAMS User Conference, Berlin 1999.
[26] Pacejka, H.B.: Tyre and Vehicle Dynamics. Oxford: Butterworth-Heinemann, 2002.
[27] Pacejka, H.B., Bakker, E.: The Magic Formula Tyre Model. Proc. 1st Int. Colloquium on
Tyre Models for Vehicle Dynamic Analysis, Swets&Zeitlinger, Lisse 1993.
146
Bibliography
[28] Pankiewicz, E. and Rulka, W.: From Off-Line to Real Time Simulations by Model Reduc-
tion and Modular Vehicle Modeling. In: Proceedings of the 19th Biennial Conference on
Mechanical Vibration and Noise Chicago, Illinois, 2003.
[29] Popp, K.; Schiehlen, W.: Fahrzeugdynamik. Teubner Stuttgart 1993.
[30] Rauh, J.: Virtual Development of Ride and Handling Characteristics for Advanced Pas-
senger Cars. Vehicle System Dynamics, 2003, Vol. 40, Nos. 1-3, pp. 135-155.
[31] Reindl, N.; Rill, G.: Modikation von Integrationsverfahren fr rechenzeitoptimale Sim-
ulationen in der Fahrdynamik, Z. f. angew. Math. Mech. (ZAMM) 68 (1988) 4, S. T107-
T108.
[32] Riepl, A.; Reinalter, W.; Fruhmann, G.: Rough Road Simulation with tire model RMOD-
K and FTire. In: Proc. of the 18th IAVSD Symposium on the Dynamics of vehicles on
Roads and on Tracks. Kanagawa, Japan, 2003. Taylor & Francis, London UK.
[33] Rill, G.: Instationre Fahrzeugschwingungen bei stochastischer Erregung. Stuttgart, Univ.,
Diss., 1983.
[34] Rill, G.: The Inuence of Correlated Random Road Excitation Processes on Vehicle Vi-
bration. In: The Dynamics of Vehicles on Road and on Tracks. Ed.: Hedrik, K., Lisse:
Swets-Zeitlinger, 1984.
[35] Rill, G.: Fahrdynamik von Nutzfahrzeugen im Daimler-Benz Fahrsimulator. In: Berech-
nung im Automobilbau, VDI-Bericht 613. Dsseldorf: VDI-Verlag 1986.
[36] Rill, G.: Demands on Vehicle Modeling. In: The Dynamics of Vehicles on Road and on
Tracks. Ed.: Anderson, R.J., Lisse: Swets-Zeitlinger 1990.
[37] Rill, G.: Vehicle Modelling for Real Time Applications. RBCM - J. of the Braz. Soc.
Mechanical Sciences, Vol. XIX - No. 2 - 1997 - pp. 192-206.
[38] Rill, G.: Modeling and Dynamic Optimization of Heavy Agricultural Tractors. In: 26th
International Symposium on Automotive Technology and Automation (ISATA). Croydon:
Automotive Automation Limited 1993.
[39] Rill, G., Salg, D., Wilks, E.: Improvement of Dynamic Wheel Loads and Ride Quality of
Heavy Agricultural Tractors by Suspending Front Axles, in: Heavy Vehicles and Roads,
Ed.: Cebon, D. and Mitchell C.G.B., Thomas Telford, London 1992.
[40] Rill, G., Chucholowski, C.: Modeling Concepts for Modern Steering Systems. In: Pro-
ceedings of Multibody Dynamics 2005. Madrid, 2005.
[41] Rill, G.: A Modied Implicit Euler Algorithm for Solving Vehicle Dynamic Equations,
Multibody System Dynamics, Volume 15, Issue 1, Feb 2006, Pages 1 - 24
147
Bibliography
[42] Rill, G.: First Order Tire Dynamics. In: Proceedings of the III European Conference on
Computational Mechanics Solids, Structures and Coupled Problems in Engineering. Lis-
bon, Portugal, 5
U8 June 2006.
[43] Rill, G.: Simulation von Kraftfahrzeugen. Vieweg Verlag, Braunschweig 1994.
[44] Rill, G.; Kessing, N.; Lange, O,; Meier, J.: Leaf Spring Modeling for Real Time Appli-
cations. In: The Dynamics of Vehicles on Road and on Tracks - Extensive Summaries,
IAVSD 03, Atsugi, Kanagawa, Japan 2003.
[45] Seibert, Th.; Rill, G.: Fahrkomfortberechnungen unter Einbeziehung der Mo-
torschwingungen. In: Berechnung und Simulation im Fahrzeugbau, VDI-Bericht 1411.
Dsseldorf: VDI-Verlag 1998.
[46] www.tesis.de.
[47] Van Oosten, J.J.M. et al: Tydex Workshop: Standardisation of Data Exchange in Tyre Test-
ing and Tyre Modelling. Proc. 2nd Int. Colloquium on Tyre Models for Vehicle Dynamic
Analysis, Swets&Zeitlinger, Lisse 1997, 272-288.
[48] Weinfurter, H.; Hirschberg, W.; Hipp, E.: Entwicklung einer Strgrenkompensation fr
Nutzfahrzeuge mittels Steer-by-Wire durch Simulation. In: Berechnung und Simulation
im Fahrzeugbau, VDI-Berichte 1846, S.923-941. VDI Verlag, Dsseldorf 2004.
148
Index
Ackermann Geometry, 135
Ackermann Steering Angle, 135, 158
Aerodynamic Forces, 121
Air Resistance, 121
Air spring, 83
All Wheel Drive, 144
Anti Dive, 134
Anti Roll Bar, 148
Anti Squat, 134
Anti-Lock-Systems, 128
Anti-roll bar, 83
Axle Kinematics, 134
Axle kinematics
Double wishbone, 7
McPherson, 7
Multi-link, 7
Axle Load, 120
Axle suspension
Solid axle, 78
Twist beam, 79
Bend Angle, 142
Bend angle, 139
Brake Pitch Angle, 129
Brake Pitch Pole, 134
Camber angle, 5, 17
Camber Compensation, 147, 150
Camber slip, 44
Caster, 8, 9
Climbing Capacity, 122
Coil spring, 82
Comfort, 97
Contact point, 18
Cornering Resistance, 143, 144
Cornering stiffness, 34
Critical velocity, 157
Curvature Gradient, 140
Damping rate, 101
Disturbance-reaction problems, 108
Disturbing force lever, 8
Down Forces, 121
Downhill Capacity, 122
Drag link, 80, 81
Drive Pitch Angle, 129
Driver, 2
Driving
Maximum Acceleration, 123
Driving safety, 97
Dynamic Axle Load, 120
Dynamic force elements, 87
Dynamic Wheel Loads, 119
Eigenvalues, 154
Environment, 3
First harmonic oscillation, 87
Fourier-approximation, 88
Frequency domain, 87
Friction, 122
Front Wheel Drive, 123, 144
Generalized uid mass, 94
Grade, 120
Hydro-mount, 93
Kingpin, 7
Kingpin Angle, 8
Lateral Acceleration, 147, 158
Lateral Force, 152
Lateral Slip, 152
Leaf spring, 82, 83
i
Index
Ljapunov equation, 109
Load, 3
Maximum Acceleration, 122, 123
Maximum Deceleration, 122, 124
Natural frequency, 101
Optimal Brake Force Distribution, 126
Optimal damping, 106, 111
Chassis, 107
Wheel, 107
Optimal Drive Force Distribution, 126
Oversteer, 158
Overturning Limit, 144
Parallel Tracks, 165
Pinion, 80
Pivot pole, 135
Power Spectral Density, 166
Quarter car model, 112, 115
Rack, 80
Random Road Prole, 165
Rear Wheel Drive, 123, 144
Reference frames
Ground xed, 4
Inertial, 4
Vehicle xed, 4
Relative damping rate, 102
Ride comfort, 108
Ride safety, 108
Road, 16
Roll Axis, 150
Roll Center, 150
Roll Steer, 168
Roll Stiffness, 146
Roll Support, 147, 150
Rolling Condition, 152
Safety, 97
Side Slip Angle, 135, 159
Sky hook damper, 111
Space Requirement, 136
Spring rate, 103
Stability, 154
Stabilizer, 83
State Equation, 154
State matrix, 113
State vector, 113
Steady State Cornering, 143, 163, 168
Steer box, 80
Steering Activity, 167
Steering Angle, 140
Steering box, 81
Steering lever, 81
Steering offset, 8, 9
Steering system
Drag link steering system, 81
Lever arm, 80
Rack and pinion, 80
Steering Tendency, 151, 157
Step Steer Input, 164, 170
Suspension model, 97
Suspension spring rate, 103
System response, 87
Tilting Condition, 122
Tire
Bore torque, 11, 46
Camber angle, 17
Camber inuence, 43
Characteristics, 39
Circumferential direction, 17
Composites, 10
Contact forces, 11
Contact patch, 11
Contact point, 16
Contact point velocity, 24
Contact torques, 11
Deection, 20
Deformation velocity, 25
Development, 10
Dynamic offset, 34
Dynamic radius, 26
Dynamics, 49
Friction coefcient, 37
Lateral direction, 17
Lateral force, 11
Lateral force characteristics, 34
ii
Index
Lateral force distribution, 34
Lateral slip, 33
Lateral velocity, 25
Lift off, 113
Linear Model, 152
Loaded radius, 17, 25
Longitudinal force, 11, 32
Longitudinal force characteristics, 33
Longitudinal force distribution, 33
Longitudinal slip, 32
Longitudinal velocity, 25
Model, 39
Normal force, 11
Pneumatic trail, 34
Radial damping, 28
Radial direction, 17
Radial Stiffness, 147
Radial stiffness, 28
Rolling resistance, 11, 30
Rolling resistance coefcient, 30
Self aligning torque, 11, 34
Sliding velocity, 34
Static radius, 17, 25, 27
Tilting torque, 11
Track normal, 17, 19
Transport velocity, 26
Tread deection, 31
Tread particles, 31
Unloaded radius, 25
Vertical force, 27
Wheel load inuence, 36
Tire Model
Kinematic, 135
Linear, 160
TMeasy, 39
Toe angle, 4
Toe-in, 4
Toe-out, 4
Torsion bar, 82
Track, 16
Track Curvature, 140
Track Radius, 140
Track Width, 135, 147
Tracknormal, 4
Trailer, 138, 141
Understeer, 158
Vehicle, 2
Vehicle comfort, 97
Vehicle dynamics, 1
Vehicle Model, 119, 129, 138, 147, 151
Vehicle model, 97, 115
Vertical dynamics, 97
Virtual Work, 147
Waviness, 167
Wheel Base, 135
Wheel camber, 5
Wheel load, 11
Wheel Loads, 119
Wheel rotation axis, 4
Wheel Suspension
Semi-Trailing Arm, 170
Single Wishbone, 170
Trailing Arm, 170
Wheel suspension
Central control arm, 79
Double wishbone, 78
McPherson, 78
Multi-Link, 78
Semi-trailing arm, 79
SLA, 79
Yaw Angle, 141
Yaw angle, 138
Yaw Velocity, 152
iii

Vehicle Dynamics

Uploaded by

Document Information

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

Vehicle Dynamics

Uploaded by

Copyright:

Available Formats

V

Figure 2.5: Power spectral density in a nite interval

function rand was used to

does not lie on the track. The corresponding track point P

= P the calculations may be repeated until the difference

full contact partial contact

Figure 3.14: Tipping torque at full contact

Figure 3.29: Two-dimensional tire characteristics

The time derivative can be transformed to a space derivative

can be converted to an equivalent lateral slip s

at the front wheels.

F = 2fdA and tan or = 90

. Hence, the phase angle which is also called

to a pure spring behavior, c

= 0, and the vehicle can be represented by two uncoupled two

= 1. In this case the viscous damping rate is more sensitive.

= 200. To approximate the lumped mass model by two decoupled two

denotes the state vector of the wheel model.

= Re()/ [[. evaluates the damping ratio.

for different values of the damping parameter d

2 0.7 for the chassis motion could

2 34 300 350 = 4900 N/(m/s) ,

, in order to let the vehicle drive a circle

and at the fully occupied one with = 135

. The steady state

UJuly 2005, 413

You might also like