Vehicle Dynamics

VEHICLE DYNAMICS
FACHHOCHSCHULE REGENSBURG
UNIVERSITY OF APPLIED SCIENCES
HOCHSCHULE FR
TECHNIK
WIRTSCHAFT
SOZIALES
LECTURE NOTES
Prof. Dr. Georg Rill
October 2005
download: http://homepages.fh-regensburg.de/%7Erig39165/
Contents
Contents
1
Introduction
1.1
Literature
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2
Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1
Vehicle Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2
Driver
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.3
Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.4
Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.5
Environment
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3
Denitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.1
Reference frames
1.3.2
Toe and camber angle
1.3.3
Design Position of Wheel Rotation Axis
1.3.4
Steering Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.4.1
Kingpin . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.4.2
Caster and Kingpin Angle . . . . . . . . . . . . . . . . . .
1.3.4.3
Caster, Steering Oset and Disturbing Force Lever
. . . .
Road
10
2.1
Modeling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
2.2
2.3
Deterministic Proles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.2.1
Bumps and Potholes
. . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.2.2
Sine Waves
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
2.3.1
Random Proles
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
Tire
3.1
3.2
3.3
3.4
19
3.1.1
Tire Development . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
3.1.2
Tire Composites
19
3.1.3
Tire Forces and Torques
. . . . . . . . . . . . . . . . . . . . . . . .
20
3.1.4
Measuring Tire Forces and Torques . . . . . . . . . . . . . . . . . .
21
Contact Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
3.2.1
Basic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
3.2.2
Local Track Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . .
25
3.2.3
Tire Deection
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
26
3.2.4
Length of Contact Patch . . . . . . . . . . . . . . . . . . . . . . . .
28
3.2.5
Static Contact Point
. . . . . . . . . . . . . . . . . . . . . . . . . .
29
3.2.6
Contact Point Velocity . . . . . . . . . . . . . . . . . . . . . . . . .
31
3.2.7
Dynamic Rolling Radius
. . . . . . . . . . . . . . . . . . . . . . . .
32
Forces and Torques caused by Pressure Distribution . . . . . . . . . . . . .
34
3.3.1
Wheel Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
34
3.3.2
Tipping Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35
3.3.3
Rolling Resistance
. . . . . . . . . . . . . . . . . . . . . . . . . . .
36
. . . . . . . . . . . . . . . . . . . . . . . . . .
37
3.4.1
Friction Forces and Torques
Longitudinal Force and Longitudinal Slip . . . . . . . . . . . . . . .
37
3.4.2
Lateral Slip, Lateral Force and Self Aligning Torque . . . . . . . . .
40
3.4.3
Wheel Load Inuence . . . . . . . . . . . . . . . . . . . . . . . . . .
41
3.4.4
Two-Dimensional Tire Characteristics . . . . . . . . . . . . . . . . .
43
3.4.5
Dierent Friction Coecients
. . . . . . . . . . . . . . . . . . . . .
45
3.4.6
Self Aligning Torque
. . . . . . . . . . . . . . . . . . . . . . . . . .
46
3.4.7
Camber Inuence . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
3.4.8
Bore Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
3.4.9
Typical Tire Characteristics
52
. . . . . . . . . . . . . . . . . . . . . .
Suspension System
54
4.1
Purpose and Components
. . . . . . . . . . . . . . . . . . . . . . . . . . .
54
4.2
Some Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
4.2.1
Multi Purpose Systems . . . . . . . . . . . . . . . . . . . . . . . . .
55
4.2.2
Specic Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
56
4.3
4.4
II
19
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Steering Systems
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
56
4.3.1
Requirements
4.3.2
Rack and Pinion Steering
56
4.3.3
Lever Arm Steering System
4.3.4
4.3.5
Bus Steer System . . . . . . . . . . . . . . . . . . . . . . . . . . . .
58
. . . . . . . . . . . . . . . . . . . . . . .
57
. . . . . . . . . . . . . . . . . . . . . .
57
Drag Link Steering System . . . . . . . . . . . . . . . . . . . . . . .
58
Standard Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .
59
4.4.1
Springs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
59
4.4.2
Damper
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
60
4.4.3
Rubber Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .
62
4.5
63
Testing and Evaluating Procedures
. . . . . . . . . . . . . . . . . .
63
4.5.2
Simple Spring Damper Combination
. . . . . . . . . . . . . . . . .
66
4.5.3
General Dynamic Force Model . . . . . . . . . . . . . . . . . . . . .
68
4.5.3.1
69
Hydro-Mount . . . . . . . . . . . . . . . . . . . . . . . . .
Vertical Dynamics
72
5.1
Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
72
5.2
Modelling Aspects
72
5.3
5.4
5.5
Dynamic Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.5.1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.1
Full Vehicle Model
5.2.2
Twodimensional Models
. . . . . . . . . . . . . . . . . . . . . . . .
73
5.2.3
Simple Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
75
Basic Tuning
. . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3.1
Natural Frequency and Damping Rate
5.3.2
Spring Rates
6.2
6.3
6.4
76
. . . . . . . . . . . . . . . .
76
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
78
5.3.2.1
Minimum Spring Rates . . . . . . . . . . . . . . . . . . . .
78
5.3.2.2
Nonlinear Springs . . . . . . . . . . . . . . . . . . . . . . .
79
5.3.3
Inuence of Damping . . . . . . . . . . . . . . . . . . . . . . . . . .
81
5.3.4
Optimal Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . .
81
5.3.4.1
Avoiding Overshoots . . . . . . . . . . . . . . . . . . . . .
81
5.3.4.2
Fast Approach to Steady State
. . . . . . . . . . . . . . .
82
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
87
Sky Hook Damper

5.4.1
Modelling Aspects
. . . . . . . . . . . . . . . . . . . . . . . . . . .
87
5.4.2
Eigenfrequencies and Damping Ratios . . . . . . . . . . . . . . . . .
88
5.4.3
Technical Realization . . . . . . . . . . . . . . . . . . . . . . . . . .
90
Nonlinear Force Elements
. . . . . . . . . . . . . . . . . . . . . . . . . . .
90
5.5.1
Quarter Car Model . . . . . . . . . . . . . . . . . . . . . . . . . . .
90
5.5.2
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
92
Longitudinal Dynamics
6.1
72
94
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
94
6.1.1
Dynamic Wheel Loads
Simple Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . .
94
6.1.2
Inuence of Grade
. . . . . . . . . . . . . . . . . . . . . . . . . . .
95
6.1.3
Aerodynamic Forces
. . . . . . . . . . . . . . . . . . . . . . . . . .
96
Maximum Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
6.2.1
Tilting Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
6.2.2
Friction Limits
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
98
6.3.1
Single Axle Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . .
98
6.3.2
Braking at Single Axle . . . . . . . . . . . . . . . . . . . . . . . . .
99
6.3.3
Optimal Distribution of Drive and Brake Forces . . . . . . . . . . . 100
6.3.4
Dierent Distributions of Brake Forces
6.3.5
Anti-Lock-Systems
Driving and Braking
Drive and Brake Pitch
. . . . . . . . . . . . . . . . 102
. . . . . . . . . . . . . . . . . . . . . . . . . . . 102
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
III
6.4.1
Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.4.2
Equations of Motion
6.4.3
Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.4.4
Driving and Braking
6.4.5
Brake Pitch Pole
7.2
7.3
109
Kinematic Approach
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.1
Kinematic Tire Model
. . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.2
Ackermann Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.3
Space Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7.1.4
Vehicle Model with Trailer . . . . . . . . . . . . . . . . . . . . . . . 112

7.1.4.1
Kinematics
7.1.4.2
Vehicle Motion
. . . . . . . . . . . . . . . . . . . . . . . . . . 112
7.1.4.3
Entering a Curve . . . . . . . . . . . . . . . . . . . . . . . 114
7.1.4.4
Trailer Motions . . . . . . . . . . . . . . . . . . . . . . . . 115
7.1.4.5
Course Calculations
. . . . . . . . . . . . . . . . . . . . . 116
Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Cornering Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.2.2
Overturning Limit
7.2.3
Roll Support and Camber Compensation . . . . . . . . . . . . . . . 121
7.2.4
Roll Center and Roll Axis
7.2.5
Wheel Loads
. . . . . . . . . . . . . . . . . . . . . . . . . . . 118
. . . . . . . . . . . . . . . . . . . . . . . 124
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Simple Handling Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

7.3.1
Modeling Concept
7.3.2
Kinematics
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.3
Tire Forces
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.3.4
Lateral Slips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.3.5
Equations of Motion
7.3.6
Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.3.8
. . . . . . . . . . . . . . . . . . . . . . . . . . . 125
. . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.3.6.1
Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.3.6.2
Low Speed Approximation . . . . . . . . . . . . . . . . . . 129
7.3.6.3
High Speed Approximation
. . . . . . . . . . . . . . . . . 129
Steady State Solution . . . . . . . . . . . . . . . . . . . . . . . . . . 130

7.3.7.1
Side Slip Angle and Yaw Velocity . . . . . . . . . . . . . . 130
7.3.7.2
Steering Tendency
7.3.7.3
Slip Angles
. . . . . . . . . . . . . . . . . . . . . . 132
. . . . . . . . . . . . . . . . . . . . . . . . . . 133
Inuence of Wheel Load on Cornering Stiness
Driving Behavior of Single Vehicles

8.1
. . . . . . . . . . . . . . . . . . . . . . . . 113
7.2.1
7.3.7
IV
. . . . . . . . . . . . . . . . . . . . . . . . . . 107
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Lateral Dynamics
7.1
. . . . . . . . . . . . . . . . . . . . . . . . . . 105
. . . . . . . . . . . 133
136
Standard Driving Maneuvers . . . . . . . . . . . . . . . . . . . . . . . . . . 136

8.1.1
Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . 136
8.1.2
Step Steer Input
8.1.3
Driving Straight Ahead . . . . . . . . . . . . . . . . . . . . . . . . . 138
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
8.2
8.3
8.1.3.1
Random Road Prole
8.1.3.2
Steering Activity . . . . . . . . . . . . . . . . . . . . . . . 140
. . . . . . . . . . . . . . . . . . . . 138
Coach with dierent Loading Conditions
. . . . . . . . . . . . . . . . . . . 140
8.2.1
Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
8.2.2
Roll Steering
8.2.3
Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . 141
8.2.4
Step Steer Input
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Dierent Rear Axle Concepts for a Passenger Car
. . . . . . . . . . . . . . 143
1 Introduction
1.1 Literature
ATZ: Automobiltechnische Zeitschrift

Fachbuchgruppe Fahrwerktechnik:
Jrnsen Reimpell, Hrsg. Vogel Buchverlag Wrzburg.
Dynamik der Kraftfahrzeuge: M. Mitschke, Bde. A,B,C; Springer-Verlag.

Mitschke,
M.;
Wallentowitz,
H.: Dynamik der Kraftfahrzeuge. 4. Auage.
Springer-Verlag Berlin Heidelberg 2004.
Popp, K.; Schiehlen, W.: Fahrzeugdynamik. Teubner Stuttgart 1993.
Simulation von Kraftfahrzeugen: G. Rill, Vieweg-Verlag 1994.
Radfhrungen der Straenfahrzeuge: W. Matschinsky, Springer-Verlag.
Blundell, M.; Harty, D.: The Multibody System Approach to Vehicle Dynamics.
Elsevier Butterworth-Heinemann Publications, 2004.
Fundamentals of Vehicle Dynamics: Th., D. Gillespie, SAE, Inc.
ISO-Standards: (International Organisation for Standardization.)
z.B.: ISO 4138 Steady State Circular Test Procedure.
Kraftfahrtechnisches
Handbuch:
Robert
Bosch
GmbH
(Hrsg.),
23.
Au.,
Vieweg-Verlag.
Proceedings:
VDI-Tagungen: z.B.: Berechnung im Automobilbau.

SAE-Congress:
FISITA:
IAVSD:
(Society of Automotive Engineers).
( Fd. Internat. des Socits d'Ingnieurs de Techniques de l'Automobile).

(International Assosiation for Vehicle System Dynamics).
Vehicle Dynamics
FH Regensburg, University of Applied Sciences
1.2 Terminology
1.2.1 Vehicle Dynamics
The expression `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.
1.2.2 Driver
By various means the driver can interfere with the vehicle:
driver
steering wheel
accelerator pedal
brake pedal
clutch
gear shift
lateral dynamics
longitudinal dynamics
vehicle
The vehicle provides the driver with these information:
vehicle
vibrations:
sounds:
motor, aerodynamics, tires
instruments:
velocity, external temperature, ...
longitudinal, lateral, vertical
driver
Prof. Dr.-Ing. G. Rill
The environment also inuences the driver:
environment
climate
track
trac density
driver
The driver's 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 dicult, if eld tests are made with test
drivers. Field tests with normal drivers have to be evaluated statistically. Of course, the
driver's 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.
1.2.3 Vehicle
The following vehicles are listed in the ISO 3833 directive:
motorcycles
passenger cars
busses
trucks
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.
Vehicle Dynamics
1.2.4 Load
Trucks are conceived for taking up load. Thus, their driving behavior changes.

Load
mass, inertia, center of gravity

dynamic behaviour (liquid load)
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 eort. When
carrying out experiments with tank trucks, ammable liquids have to be substituted with
water. Thus, the results achieved cannot be simply transferred to real loads.
1.2.5 Environment
The environment inuences primarily the vehicle:

Environment
road:
irregularities, coecient of friction
air:
resistance, cross wind

vehicle
but also aects the driver:

environment
climate
visibility

driver
Through the interactions between vehicle and road, roads can quickly be destroyed.
The greatest diculty 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.
1.3 Definitions
1.3.1 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
x0 , y0 , z0 serves as an inertial reference frame. Within the vehicle-xed

reference frame the xF -axis points forward, the yF -axis to the left, and the zF -axis upward.
with the axis
The wheel rotates around an axis which is xed to the wheel carrier. The reference frame
is xed to the wheel carrier. In design position its axes
the corresponding axis of vehicle-xed reference frame
F.
xC , yC
and
zC
are parallel to
zF
yF
xF
z0
zC
x0
yC
en
xC
y0
eyR
Figure 1.1: Frames used in vehicle dynamics
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
eyR
into the direction of
the wheel rotation axis.

Finally, the normal vector
en
describes the inclination of the local track plane.
1.3.2 Toe and camber angle

front
xF
left
wheel
yF
vehicle
center
plane
right
wheel
rear
Figure 1.2: Positive toe-in angle
According to the DIN 70 000 directive the angle
between the vehicle center plane in
longitudinal direction and the intersection line of the tire center plane with the track
plane is named toe or toe-in angle. It will be positive, if the front part of the wheel is
oriented towards the vehicle center plane, Figure 1.2. Toe-in reduces the tendency of the
wheels to shimmy.
The camber angle
en .
is the angle between the wheel center plane and the local track normal
It will be positive, if the upper part of the wheel is inclined outwards, Figure 1.3. A
cambered wheel causes a non symmetric tire wear.
Vehicle Dynamics
en
en
left
wheel
right
wheel
zF
yF
road
Figure 1.3: Positive camber angle
1.3.3 Design Position of Wheel Rotation Axis

The unit vector
eyR
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 design position the corresponding axes of the frames
and
are parallel.
zC = zF
xC = xF
0
0
0*
yC = yF
eyR
Figure 1.4: Design position of wheel rotation axis

Then, for the left wheel we get
eyR,F = eyR,C
tan 0
1
1
= q
tan2 0 + 1 + tan2 0
tan 0
or
eyR,F = eyR,C
(1.2)
yF -axis and the projection line of the wheel rotation

0 describes the angle between the yF -axis and the
0
projection line of the wheel rotation axis into the yF - zF -plane, whereas 0 is the angle
between the wheel rotation axis eyR and its projection into the xF - yF -plane. Kinematics
where
sin 0 cos 0
= cos 0 cos 0 ,
sin 0
(1.1)
is the angle between the
axis into the
xF - yF -plane,
the angle
and compliance test machines usually measure the angle

industry mostly uses this angle instead of
That is why, the automotive
en
points into the direction of
correspond with the toe angle and the
camber angle 0 . To specify the dierence between 0 and 0 the ratio between the third
the vertical axes
the angles
0 .
0 .
On a at and horizontal road where the track normal
zC = zF
and second component of the unit vector

deliver
and
eyR
is considered. The Equations 1.1 and 1.2
sin 0
tan 0
=
1
cos 0 cos 0
Hence, for small angles
0 1
or
tan 0 =
tan 0
.
cos 0
the dierence between the angles
(1.3)
and
is hardly
noticeable.
1.3.4 Steering Geometry

1.3.4.1 Kingpin
At the steered front axle, 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, Fig. 1.5 and Fig. 1.6.
zR
B
yR
M
xR
kingpin axis A-B

Figure 1.5: Double wishbone wheel suspension
The wheel body rotates around the kingpin at steering motions. At the double wishbone
B , which determine the kingpin, are both xed to the wheel

A is still xed to the wheel body at the standard McPherson
wheel suspension, the ball joint B is now xed to the vehicle body. At a multi-link axle the
axis the ball joints
and
body. Whereas the ball joint
kingpin is no longer dened by real joints. Here, as well as with the enhanced McPherson
wheel suspension, the kingpin changes its position relative to the wheel body at wheel
travel and steering motions.
Vehicle Dynamics
B
zR
zR
yR
yR
xR
xR
M
A
rotation axis
kingpin axis A-B
Figure 1.6: McPherson and multi-link wheel suspensions
1.3.4.2 Caster and Kingpin Angle
eS describes the direction of the kingpin axis. Within the vehicle xed
reference frame F it can be xed by two angles. The caster angle denotes the angle
between the zF -axis and the projection line of eS into the xF -, zF -plane. In a similar
way the projection of eS into the yF -, zF -plane delivers the kingpin inclination angle ,
The unit vector
Fig. 1.7.
eS
zF
zF
yF
xF
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 deliver a virtual kingpin. The current values of the caster angle
inclination angle
and the kingpin
can be calculated from the components of the unit vector
eS
in the
direction of the kingpin, described in the vehicle xed reference frame
(1)
tan =
eS,F
(3)
eS,F
where
(1)
(2)
(3)
eS,F , eS,F , eS,F
xed reference frame
(2)
and
tan =
eS,F
are the components of the unit vector
F.
(1.4)
eS,F
expressed in the vehicle
(3)
eS,F
1.3.4.3 Caster, Steering Offset and Disturbing Force Lever

The contact point
P,
the local track normal
en
and the unit vectors
ex
and
ey
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
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
to
can be written as
rSP = c ex + s ey ,
where
names the caster and
positive, if
is the steering oset.
is located in front of and inwards of
(1.5)
Caster and steering oset will be
P.
kingpin line
C d
ey
en
P
ex
S
Figure 1.8: Caster and steering oset

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.
2 Road
2.1 Modeling Aspects
Sophisticated road models provide the road height
at each point
x, y ,
zR
and the local friction coecient
Fig. 2.1.
Road profile
z0
y0
Segments
z(x,y)
x0
(x,y)
Friction
Obstacle
Center Line L(s)
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.
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 eort 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
z
x
z2
z1
zR(x,y)
z1(s)
Figure 2.2: Parallel track road model
Now, the road heights on the left and right track are provided by two one-dimensional
functions
z1 = z1 (s)
z2 = z2 (s).
and
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
z1 (s) z2 (s).
approximately the same,
z1 (x) = z2 (x)
Then, a single track road model with
zR (s) =
can be used. Now, the roll excitation of the vehicle is neglected too.
2.2 Deterministic Profiles

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.
H
B
y
L
Figure 2.3: Rectangular cleat and cosine-shaped bump

Then, the rectangular cleat is simply dened by
(
z(x, y) =
H
0
if
else
0<x<L
and
12 B < y < 12 B
(2.1)
11
Vehicle Dynamics
and the cosine-shaped bump is given by
z(x, y) =
1
2
where
H, B
if negative
x
H 1 cos 2
L
0

0<x<L
if
and
21 B < y < 12 B
(2.2)
else
L denote height, width and length of the obstacle. Potholes are obtained
values for the height (H < 0) are used.
and
2.2.2 Sine Waves

Using the parallel track road model, a periodic excitation can be realized by
z2 (s) = A sin ( s ) ,
z1 (s) = A sin ( s) ,
(2.3)
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 z1 = z2 and the out of phase excitation with
z1 = z2 .
where
ds/dt = v0 , the momentary position of the

initial position s = 0 at t = 0 was assumed. By
If the vehicle runs with constant velocity

vehicle is given by
s = v0 t,
where the
introducing the wavelength
(2.4)
2
2
v0
s=
v0 t = 2 t = t .
L
L
L
(2.5)
L =
the term
can be written as
s =
Hence, in the time domain the excitation frequency is given by
f = /(2) = v0 /L.
0.5 Hz to 15 Hz . This
v0 /L 0.5 Hz and v0 /L 15 Hz .
For most of the vehicles the rigid body vibrations are in between
range is covered by waves which satisfy the conditions
L = 4 m, the rigid body vibration of a vehicle are excited

min
if the velocity of the vehicle will be varied from v0
= 0.5 Hz 4 m = 2 m/s = 7.2 km/h
max
= 15 Hz 4 m = 60 m/s = 216 km/h. Hence, to achieve an excitation in the
to v0
For a given wavelength, lets say
whole frequency range with moderate vehicle velocities proles with dierent varying
wavelengths are needed.
2.3 Random Profiles

2.3.1 Statistical Properties
Road proles t the category of stationary Gaussian random processes. Hence, the irregularities of a road can be described either by the prole itself
properties, Fig. 2.4.
12
zR = zR (s) or by its statistical
zR
0.15
Gaussian
density
function
Realization
0.10
[m]
0.05
0
-0.05
-0.10
-0.15
-200
Histogram
s
-100
-150
-50
50
200
[m] 150
100
Figure 2.4: Road prole and statistical properties
By choosing an appropriate reference frame, a vanishing mean value
1
m = E {zR (s)} = lim
X X
ZX/2
zR (s) ds = 0
(2.6)
X/2
can be achieved, where
E {} denotes the expectation operator. Then, the Gaussian density
function which corresponds with the histogram is given by
p(zR ) =
where the deviation or the eective value
zR2
e 2 2 ,
(2.7)
is obtained from the variance of the process
zR = zR (s)
ZX/2

1
2 = E zR2 (s) = lim
X X

zR (s)2 ds .
(2.8)
X/2
Alteration of
eects the shape of the density function. In particular, the points of
inexion occur at
The probability of a value
P () =
|z| <
is given by
Z+ z 2
e 2 2 dz .
P () = 0.683, P (2) = 0.955,

|z| 3 is 0.3%.
In particular, one gets the values:

Hence, the probability of a value
(2.9)
and
P (3) = 0.997.
In extension to the variance of a random process the auto-correlation function is dened

by
1
R() = E {zR (s) zR (s+)} = lim
X X
ZX/2
zR (s) zR (s+) ds .
(2.10)
X/2
13
Vehicle Dynamics
R() = R(),
The auto-correlation function is symmetric,
and it plays an important
part in the stochastic analysis. In any normal random process, as
increases the link
zR (s) and zR (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
2
coincides with the variance of the process. If a periodic term is present in the process
it will show up in R().
between
Usually, road proles are characterized in the frequency domain. Here, the auto-correlation
function
S()
R()
is replaced by the power spectral density (psd)
S().
In general,
R()
and
are related to each other by the Fourier transformation
1
S() =
2
R() e
1
R() =
2
and
where
S() ei d ,
(2.11)
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() ,
the relationship
if
() = 0 ,
and
ei = cos() i sin(),
if
<0,
and the symmetry property
(2.12)
R() = R()
Eq. (2.11) results in
2
() =
Z
R() cos () d
and
Z
R() = () cos () d .
(2.13)
Now, the variance is obtained from
Z
= R( = 0) = () d .
2
(2.14)
0
In reality the psd
()
will be given in a nite interval
1 N ,
Fig. 2.5. Then,
(i)
Figure 2.5: Power spectral density in a nite interval

Eq. (2.14) can be approximated by a sum, which for
N
X
i=1
14
(i ) 4
with
equal intervals will result in
4 =
N 1
.
N
(2.15)
2.3.2 Classification of Random Road Profiles

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 )
where,
= 2/L
in
rad/m
w
describes the value of the psd at a the reference wave
Power spectral density [m2/(rad/m)]
(2.16)
0 = (0 ) in m2 /(rad/m)
number 0 = 1 rad/m. The drop
denotes the wave number and
in magnitude is modeled by the waviness
10-3
w.
a) Measurements (country road)
b) Range of road classes (ISO 8608)

0=256106
10-4
10-5
Class E
0=1106
10-6
10-7
10-8
10-9 -2
10
Class A
10-1
100
101
Wave number [rad/m]
102
10-2
10-1
100
101
Wave number [rad/m]
102
Figure 2.6: Road power spectral densities: a) Measurements, b) Classication

According to the international directive ISO 8608 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
6
its reference value 0 . Class A with 0 = 1 10
m2 /(rad/m) characterizes very smooth
6
highways, whereas Class E with 0 = 256 10
m2 /(rad/m) represents rather rough
roads, Fig. 2.6b.
15
Vehicle Dynamics
2.3.3 Realizations
2.3.3.1 Sinusoidal Approximation
A random prole of a single track can be approximated by a superposition of
sine waves
zR (s) =
N
X
Ai sin (i s i ) ,
(2.17)
i=1
Ai and its
i , i = 1(1)N
i .
where each sine wave is determined by its amplitude
wave number
dierent sets of uniformly distributed phase angles
in the range between
and
By
dierent proles can be generated which are similar in the general appearance
but dierent in details.

The variance of the sinusoidal representation is then given by
1
2 = lim
X X
ZX/2 X
N
X/2
For
i=j
and for
i 6= j
!
Ai sin (i s i )
i=1
N
X
!
Aj sin (j s j )
ds .
(2.18)
j=1
dierent types of integrals are obtained. The ones for
i=j
can
be solved immediately
Z
Jii =
A2i

1
A2i
i si sin 2 (i si ) .
sin (i si ) ds =
2i
2
2
(2.19)
Using the trigonometric relationship
sin x sin y =
the integrals for
i 6= j
1
1
cos(xy) cos(x+y)
2
2
(2.20)
can be solved too
Z
Jij =
Ai sin (i si ) Aj sin (j sj ) ds
1
= A i Aj
2
=
1
cos (ij s ij ) ds Ai Aj
2
Z
cos (i+j s i+j ) ds
(2.21)
1 Ai Aj
1 A i Aj
sin (ij s ij ) +
sin (i+j s i+j )
2 ij
2 i+j
where the abbreviations
ij = ij
and
ij = ij
terms in Eqs. (2.19) and (2.21) are limited to values of
were used. The sine and cosine
1.
Hence, Eq. (2.18) simply
results in
N
N
N
1 X X/2
1 X X/2
1X 2
Jii X/2 + lim
Jij X/2 =
= lim
A .
X X
X X
2 i=1 i
i=1
i,j=1
|
{z
}
|
{z
}
N
2
0
X A
i
i
2
i
i=1
2
16
(2.22)
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
given psd
()
described by Eq. (2.17) will have a
if the amplitudes are generated according to
Ai =
and the wave numbers
2 (i ) 4 ,
are chosen to lie at
i = 1(1)N ,
equal intervals
(2.23)
4.
Road profile z=z(s)
0.10
[m]
zR = zR (s)
0.05
0
-0.05
-0.10
0
10
20
30
40
50
60
70
80 [m] 90
100
Figure 2.7: Realization of a country road
0 = 10 106 m2 /(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
Ai , i = 1(1)N were calculated by Eq. (2.23) and the MATLABr function rand was used
to produce uniformly distributed random phase angles in the range between 0 and 2 .
A realization of the country road with a psd of
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. A simple rst order shaping lter for the road prole
where
is a constant, and
d
zR (s) = zR (s) + w(s) ,
ds
w(s) is a white noise process with
zR
reads as
(2.24)
the spectral density
w .
Then, the spectral density of the road prole is obtained from
R = H() W H T () =
where
is the wave number, and
H()
1
W
1
W
= 2
,
+ i
i
+ 2
(2.25)
is the frequency response function of the shaping
lter.
By setting
W = 10 106 m2 /(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. Here, the crosscorrelation between the irregularities of the left and right track have to be taken into
account too.
17
Power spectral density [m2/(rad/m)]
Vehicle Dynamics
10-3
Measurements
Shaping filter
10-4
10-5
10-6
10-7
10-8
10-9 -2
10
10-1
100
101
Wave number [rad/m]
102
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
zR = zR (x, y)
via a sinusoidal ap-
proximation 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 gener-
ator a reproducible road prole can be generated.
z
x
1
0
-1
4
0
-2
-4
10
15
20
25
30
35
40
45
50
Figure 2.9: Two-dimensional road prole

By assuming the same statistical properties in longitudinal and lateral direction twodimensional 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 tires are shown in Table 3.1.
1839
1845
Charles Goodyear: vulcanization

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 separate components into a cohesive product that satises the driver's 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
Vehicle Dynamics
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 tire's prole design the contact patch forms a not
necessarily coherent area, Figure 3.1.
140 mm
180 mm
Figure 3.1: Tire footprint of a passenger car at normal loading condition:

Continental 205/55 R16 90 H,
2.5
bar,
Fz = 4700 N
The eect 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
to the
z -axis
z -axis
is normal to the track, the
and perpendicular to the wheel rotation axis
right-handed reference frame also xes the
eyR .
x-axis
is perpendicular
Then, the demand for a
y -axis.
The components of the contact force vector are named according to the direction of the
axes, Figure 3.2.
A non symmetric distribution of the forces in the contact patch causes torques around
the
and
axes. A cambered tire generates a tilting torque
Tx .
The torque
the rolling resistance of the tire. In particular, the torque around the
z -axis
Ty
includes
is important
in vehicle dynamics. It consists of two parts,
Tz = TB + TS .
20
(3.1)
Fx
Fy
Fz
longitudinal force
Tx
Ty
Tz
tilting torque
lateral force
vertical force or wheel load
eyR
Fy
rolling resistance torque
Tx
Fx
self aligning and bore torque
Tz
Ty
Fz
Figure 3.2: Contact forces and torques
The rotation of the tire around the

torque
MS
z -axis
causes the bore torque
MB .
The self aligning
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, Figure
3.4. Here, the measurements are performed under real operating conditions. Arbitrary
Test trailer
tire
exact contact
real road
compensation wheel
test wheel
Figure 3.3: Layout of a tire test trailer

surfaces like asphalt or concrete and dierent 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, Figure 3.4.
21
Vehicle Dynamics
tire
too small
contact area
safety walk
coating
tire
safety walk coating
perfect contact
rotation
drum
tire
too large contact area
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.
Radial 205/50 R15, FN= 3500 N, dry asphalt
4000
Longitud force Fx [N]
3000
2000
Driving
1000
0
-1000
Braking
-2000
-3000
-4000
-40
-30
-20
-10
10
20
30
40
Longitudinal slip [%]
Figure 3.5: Typical results of tire measurements
Tire forces and torques are measured in quasi-static operating conditions. Hence, the measurements for increasing and decreasing the sliding conditions usually result in dierent
graphs, Figure 3.5. In general, the mean values are taken as steady state results.
22
3.2 Contact Geometry

3.2.1 Basic Approach
x0 -, y0 - z0 -system
The current position of a wheel in relation to the xed

wheel center
and the unit vector
eyR
is given by the
in the direction of the wheel rotation axis, Figure
3.6.
rim
centre
plane e zR
tire
e yR
rMP
e yR
en
wheel
carrier
ex
b
P0 a
ey P
en
P0
z0
P*
local road plane
x0
road: z = z ( x , y )
x0
y0
y0
z0
Figure 3.6: Contact geometry
The irregularities of the track can be described by an arbitrary function of two spatial
coordinates
z = z(x, y).
At an uneven track the contact point
(3.2)
can not be calculated directly. At rst, one can
get an estimated value with the vector
rM P = r0 ezB ,
where
r0
is the undeformed tire radius, and
ezB
(3.3)
is the unit vector in the
z -direction of the
body xed reference frame.

The position of this rst guess
z0
with respect to the earth xed reference frame
x0 , y0 ,
is determined by
x
= y ,
z
r0P ,0 = r0M,0 + rM P ,0
(3.4)
23
Vehicle Dynamics
where the vector
r0M
describes the position of the rim center
P0
does not lie on the track. The corresponding track point
r0P0 ,0 =
M.
Usually, the point
follows from
(3.5)
z (x , y )
where Eq. (3.2) was used to calculate the appropriate road height. In the point
track normal
en
P0
the
is calculated, now. Then the unit vectors in the tire's circumferential
direction and lateral direction can be determined. One gets
ex =
where
ex
eyR en
| eyR en |
ey = en ex ,
and
(3.6)
eyR denotes the unit vector into the direction of the wheel rotation axis. Calculating
eyR not always being perpendicular to the track. The tire
demands a normalization, as
camber angle
= arcsin eTyR en
(3.7)
describes the inclination of the wheel rotation axis against the track normal.
The vector from the rim center
P0
to the track point
is split into three parts now
rM P0 = rS ezR + a ex + b ey ,
where
rS
denotes the loaded or static tire radius,
a, b
(3.8)
are distances measured in circum-
ferential and lateral direction, and the radial direction is given by the unit vector
ezR = ex eyR
which is perpendicular to
ex
and
eyR .
(3.9)
A scalar multiplication of Eq. (3.8) with
en
results
in
eTn rM P0 = rS eTn ezR + a eTn ex + b eTn ey .

As the unit vectors
ex
and
ey
are perpendicular to
en
(3.10)
Eq. (3.10) simplies to
eTn rM P0 = rS eTn ezR .
(3.11)
Hence, the static tire radius is given by
rS
The contact point
eTn rM P0
= T
.
en ezR
(3.12)
given by the vector
rM P = rS ezR
(3.13)
lies within the rim center plane. The transition from the point
takes place according to Eq. (3.8) by the terms
24
a ex
and
b ey
P0
to the contact point
perpendicular to the track
normal
en .
The track normal, however, was calculated in the point
track the point
P0 .
With an uneven
no longer lies on the track and can therefor no longer considered as
contact point.
P = P now the Eqs. (3.5)

P and P0 is suciently small.
With the newly estimated value

until the dierence between
to (3.13) can be repeated
Tire models which can be simulated within acceptable time assume that the contact patch
is suciently at. At an ordinary passenger car tire, the contact area has at normal load
approximately the size of
1520cm. 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 Local Track Plane

A plane is given by three points. In order to get a good approximation to the local track
unevenness four point will be used to determine the local track normal. Using the initial
guess in Eq. (3.3) and the wheel rotation axis
estimated by
ex =
eyr
the circumferential direction can be
eyR ezB
.
| eyR ezB |
(3.14)
Similar to Eq. (3.4) four points are generated now
xi
= yi ,
zi
i = 1(1)4 .
(3.15)
In order to sample the contact patch as good as possible the tire width
b and the unloaded
r0Qi ,0 = r0M,0 + rM Qi ,0
tire radius
r0
are used to place the points via
x r0 ex
r0 ezB ,
rM Q2 = x r0 ex
r0 ezB ,
rM Q1 =
y b eyR r0 ezB ,
rM Q3 =
(3.16)
y b eyR r0 ezB
rM Q4 =
in the front, in the rear, in the left, and in the right of the contact patch.
According to Eq. (3.5) the corresponding points on the track are given by
r0Qi ,0
xi
=
yi
,
z (xi , yi )
i = 1(1)4 .
(3.17)
The calculation of the track normal is straight forward now, Figure 3.7. The vectors
25
Vehicle Dynamics
eyR
Q1*
rMP*
en
Q1
Q3*
Q3
P*
rQ2Q1
P
Q4*
Q4
Q2*
rQ3Q4
Q2
Figure 3.7: Local track plane
rQ2 Q1 = r0Q1 r0Q2
and
rQ4 Q3 = r0Q3 r0Q4
dene the local track inclination in longitudinal
and lateral direction. Hence, the local track normal is dened by
en =
The unit vectors
ex , ey
rQ2 Q1 rQ4 Q3
.
| rQ2 Q1 rQ4 Q3 |
(3.18)
in longitudinal and lateral direction are calculated from Eq. (3.6).
The mean value of the track points
r0P0 ,0 =
serves as rst improvement of
point
1
(r0Q1 ,0 + r0Q2 ,0 + r0Q3 ,0 + r0Q4 ,0 )
4
the contact point, P P0 . Finally,
(3.19)
the corresponding
in the rim center plane is obtained by Eqs. (3.12) and (3.13). On rough roads
the point
not always is located on the track. But, together with the local track normal
it represents the local road plane very well. As in reality, sharp bends and discontinuities
which will occur at step- or ramp-sized obstacles are smoothed by this approach.
3.2.3 Tire Deflection

For a vanishing camber angle
=0
the deected zone has a rectangular shape, Figure
3.8. Its area is given by
A0 = 4z b ,
where
(3.20)
is the width of the tire, and the tire deection is obtained by
4z = r0 rS .
(3.21)
Here, the width of the tire simply equals the width of the contact zone,
wC = 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
rSL = rs +
26
b
tan
2
and
rSR = rs
b
tan
2
(3.22)
=/ 0
=0
eyR
eyR
rS
rSL
r0
en
eyR
rSR
rS
en
rS
r0
en
r0
z
b
wC = b
rSR
b*
wC
wC
full contact
partial contact
Figure 3.8: Tire deection
determines the shape of the deected zone.
rSL r0
The tire will be in full contact to the road if
and
rSR r0
hold. Then, the
deected zone has a trapezoidal shape with an area of
A =
1
(r0 rSR + r0 rSL ) b = (r0 rS ) b .
2
Equalizing the cross sections
A0 = A
(3.23)
results in
4z = r0 rS .
Hence, at full contact the tire camber angle
(3.24)
has no inuence to the vertical tire force.
But, due to
wC =
b
cos
(3.25)
the width of the contact area increases with the tire camber angle.
The deected zone will change to a triangular shape if one of the ank radii exceeds the
undeected tire radius. Assuming
rSL > r0
obtained by
A =
and
rSR < r0
1
(r0 rSR ) b ,
2
the area of the deected zone is
(3.26)
where the width of the deected zone follows from
b =
r0 rSR
.
tan
(3.27)
Now, Eq. (3.26) reads as
1 (r0 rSR )2
A =
.
2 tan
(3.28)
27
Vehicle Dynamics
Equalizing the cross sections
A0 = A
results in
1 r0 rS + 2b tan
4z =
2
b tan
where Eq. (3.22) was used to express the ank radius
tire width
and the camber angle
wC =
2
rSR
(3.29)
by the static tire radius
| tan |
r0 rS + 2b tan
r0 rSR
b
=
=
,
cos
tan cos
sin
and
| sin |
the
Now, the width of the contact area is given by
(3.30)
where the Eqs. (3.27) and (3.22) where used to simplify the expression. If
are replaced by
rS ,
tan
and
sin
then, the Eqs. (3.29) and (3.30) will hold for positive
and negative camber angles.
3.2.4 Length of Contact Patch

To approximate the length of the contact patch the tire deformation is split into two
parts, Figure 3.9. By
4zF
and
4zB
the average tire ank and the belt deformation are
measured. Hence, for a tire with full contact to the road
4z = 4zF + 4zB = r0 rS
(3.31)
will hold.
undeformed
belt
Belt
Fz
Rim
zF
rS r0
r0
L/2
zB
zB
Figure 3.9: Length of contact patch
Assuming both deections being equal will lead to
4zF 4zB
1
4z .
2
(3.32)
Approximating the belt deection by truncating a circle with the radius of the undeformed
tire results in
28
2
L
+ (r0 4zB )2 = r02 .
2
(3.33)
In normal driving situations the belt deections are small,
4zB r0 .
Hence, Eq. (3.33)
can be simplied and nally results in
L2
= 2 r0 4zB
4
L =
or
p
8 r0 4zB .
(3.34)
Inspecting the passenger car tire footprint in Figure 3.1 leads to a contact patch length of
L 140mm. For this tire the radial stiness and the inated radius are specied with cR =
265 000 N/m and r0 = 316.9 mm. The overall tire deection can be estimated by 4z =
Fz /cR . At the load of Fz = 4700N the deection amounts to 4z = 4700N / 265 000N/m =
0.0177 m. Then, by approximating the belt deformation
by the half of the tire deection,
p
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.
3.2.5 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
static contact point
Q,
FZ
will be shifted from the geometric contact point
to the
Figure 3.10.
rS
en
P
ey
r0-rSL
Fz
A
r0-rSR
y
wC
Figure 3.10: Lateral deviation of contact point at full contact

The center of the trapezoidal area determines the lateral deviation
yQ .
By splitting the
area into a rectangular and a triangular section we will obtain
yQ =
y A + y 4 A4
.
A
(3.35)
The minus sign takes into account that for positive camber angles the acting point will
move to the right whereas the unit vector
ey
dening the lateral direction points to the
left. The area of the whole cross section results from
A =
1
(r0 rSL + r0 rSR ) wC ,
2
(3.36)
29
Vehicle Dynamics
where the width of the contact area
wC
is given by Eq. (3.25). Using the Eqs. (3.22) and
(3.24) the expression can be simplied to
A = 4z wC .
(3.37)
As the center of the rectangular section is located on the center line which runs through the
geometric contact point,
y = 0
will hold. The distance from the center of the triangular
section to the center line is given by
y4 =
1
1
1
wC wC = wC .
2
3
6
(3.38)
Finally, the area of the triangular section is dened by
A4 =
1
1
1
(r0 rSR (r0 rSL )) wC =
(rSL rSR )) wC =
(b tan ) wC ,
2
2
2
(3.39)
where Eq. (3.22) was used to simplify the expression. Now, Eq. (3.35) can be written as
yQ =
1
6
wC
1
2
b tan wC
b tan
b2 tan
=
wC =
.
4z wC
12 4z
12 4z cos
(3.40)
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, Figure 3.11.
b/2
en
ey
Q
Fz
y
wC
Figure 3.11: Lateral deviation of contact point at partial contact

Now, the location of the static contact point

yQ =
where the width of the contact area
wC
is given by
1
b
wC
3
2 cos

,
(3.41)
is determined by Eq. (3.30) and the term
describes the distance from the geometric contact point
b/(2 cos )
to the outer corner of the
contact patch. The plus sign holds for positive and the minus sign for negative camber
angles.
The static contact point
described by the vector
r0Q = r0P + yQ ey
represents the contact patch much better than the geometric contact point
30
(3.42)
P.
3.2.6 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
given by Eq. (3.42) can be expressed
as follows
r0Q = r0M + rM Q ,
where
(3.43)
rM Q describes the position of

Q relative to the wheel center M . The absolute velocity of the contact
denotes the wheel center and hence, the vector
point will be obtained from
v0Q,0 = r0Q,0 = r0M,0 + rM Q,0 ,

where
r0M,0 = v0M,0
(3.44)
denotes the absolute velocity of the wheel center. The vector
rM Q
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
4z
normal to the road.
Hence, its time derivative can be calculated from
rM Q,0 = 0R,0
rM Q,0 + 4z en,0 ,
where
0R
(3.45)
is the angular velocity of the wheel rim without any component in the direction
of the wheel rotation axis,
4z
denotes the change of the tire deection, and
en
describes
the road normal. Now, Eq. (3.44) reads as
v0Q,0 = v0M,0 + 0R,0

rM Q,0 + 4z en,0 .
As the point
lies on the track,
v0Q,0
(3.46)
must not contain any component normal to the
track
eTn,0 v0P,0 = 0
As
en,0
or
is a unit vector,
eTn,0 v0M,0 + 0R,0

rM Q,0 + 4z eTn,0 en,0 = 0 .
eTn,0 en,0 = 1
(3.47)
will hold, and then, the time derivative of the tire
deformation is simply given by
4z = eTn,0 v0M,0 + 0R,0

rM Q,0 .
(3.48)
Finally, the components of the contact point velocity in longitudinal and lateral direction
are obtained from
rM Q,0
vx = eTx,0 v0Q,0 = eTx,0 v0M,0 + 0R,0
(3.49)
and
rM Q,0 ,
vy = eTy,0 v0P,0 = eTy,0 v0M,0 + 0R,0
where the relationships
(3.50)
eTx,0 en,0 = 0 and eTy,0 en,0 = 0 were used to simplify the expressions.
31
Vehicle Dynamics
deflected tire
rigid wheel
r0 r
S
rD
vt
x
Figure 3.12: Dynamic rolling radius
3.2.7 Dynamic Rolling Radius

At an angular rotation of
4, assuming the tread particles stick to the track, the deected

x, Figure 3.12.
tire moves on a distance of

With
r0
as unloaded and
rS = r0 4r
as loaded or static tire radius
r0 sin 4 = x
(3.51)
r0 cos 4 = rS
(3.52)
and
hold.
If the motion of a tire is compared to the rolling of a rigid wheel, then, its radius
have to be chosen so that at an angular rotation of
rD
will
the tire moves the distance
r0 sin 4 = x = rD 4 .
(3.53)
Hence, the dynamic tire radius is given by
rD =
For
4 0
r0 sin 4
.
4
one obtains the trivial solution
(3.54)
rD = r0 .
At small, yet nite angular rotations the sine-function can be approximated by the rst
terms of its Taylor-Expansion. Then, Eq. (3.54) reads as
rD
4 16 43
= r0
= r0
4
1
1 42
6

.
(3.55)
With the according approximation for the cosine-function
rS
1
= cos 4 = 1 42
r0
2
32
or
4 = 2
rS
1
r0

(3.56)
one nally gets

r D = r0
Due to
rS = rS (Fz )
1
1
3
the ctive radius
rS
1
r0
rD

=
2
1
r0 + rS .
3
3
Fz . Therefore,
velocity , then
depends on the wheel load
called dynamic tire radius. If the tire rotates with the angular
(3.57)
v t = rD
it is
(3.58)
will denote the average velocity at which the tread particles are transported through the
contact patch.
[mm]
10
0
-10
r -r
D
-20
Fz [kN]
Measurements
TMeasy tire model
Figure 3.13: Dynamic tire radius

In extension to Eq. (3.57), the dynamic tire radius is approximated in the tire model
TMeasy by

FzS
= r0 + (1 ) r0
c
|
{z 0 }
rS

rD
(3.59)
where the static tire radius rS = r0 4r has been approximated by using the linearized
S
tire deformation 4r = Fz /c0 . The parameter is modeled as a function of the wheel
load Fz

= N + ( 2N N )
N and 2N
Fz = 2FzN .
where
Fz
1
FzN
denote the values for the pay load
Fz = FzN
(3.60)
and the doubled pay load
With the TMeasy parameters for a passenger car tire
vertical tire stiffness at fz=fz0

vertical tire stiffness at fz=2*fz0
[N/m],
[N/m],
coefficient for dynamic tire radius fz=fz0 [-],

coefficient for dynamic tire radius fz=2*fz0 [-],
190000.
206000.
0.375
0.750
the approximation of measured tire data can be done very well, Figure 3.13.
33
Vehicle Dynamics
3.3 Forces and Torques caused by Pressure Distribution

3.3.1 Wheel Load
The vertical tire force
Fz
can be calculated as a function of the normal tire deection
and the deection velocity
4z
4z
Fz = Fz (4z, 4z)
.
(3.61)
Because the tire can only apply pressure forces to the road the normal force is restricted
to
Fz 0.
In a rst approximation
Fz
is separated into a static and a dynamic part
Fz = FzS + FzD .
(3.62)
The static part is described as a nonlinear function of the normal tire deection
FzS = a1 4z + a2 (4z)2 .
The constants
a1
and
a2
(3.63)
may be calculated from the radial stiness at nominal and double
payload
cN

d FzS
=
d 4z FzS =FzN
c2N
and

d FzS
=
.
d 4z FzS =2FzN
(3.64)
The derivative of Eq. (3.63) results in
d FzS
= a1 + 2 a2 4z .
d 4z
(3.65)
From Eq. (3.63) one gets
4z =
a1
a21 + 4a2 FzS

.
2a2
(3.66)
Because the tire deection is always positive, the minus sign in front of the square root
has no physical meaning, and can be omitted therefore. Hence, Eq. (3.65) can be written
as
d FzS
= a1 + 2 a2
d 4z
a1 +
a21 + 4a2 FzS

2a2
!
=
a21 + 4a2 FzS .
(3.67)
Combining Eqs. (3.64) and (3.67 results in
cN
c2N
a21 + 4a2 FzN

q
a21 + 4a2 2FzN
=
or
c2N
or
c22N
and
a2 =
= a21 + 4a2 FzN ,

=
a21
(3.68)
8a2 FzN
nally leading to
a1 =
34
2 c2N c22N
c22N c2N
.
4 FzN
(3.69)
Results for a passenger car and a truck tire are shown in Figure 3.14. The parabolic
approximation in Eq. (3.63) ts very well to the measurements. The radial tire stiness
Fz = 3 200 N can be specied with c0 =

Fz = 35 000 N and the stiness c0 = 1 250 000N/m of a truck
of the passenger car tire at the payload of
190 000N/m.
The Payload
tire are signicantly larger.
Passenger Car Tire: 205/50 R15
80
60
40
20
2
0
Truck Tire: X31580 R22.5
100
Fz [kN]
Fz [kN]
10
10
20
30
z [mm]
40
50
Figure 3.14: Tire radial stiness:
20
40
60
z [mm]
80
Measurements, Approximation
The dynamic part is roughly approximated by
FzD = dR 4z ,
where
dR
(3.70)
is a constant describing the radial tire damping, and the derivative of the tire
deformation
4z
is given by Eq. (3.48).
3.3.2 Tipping Torque

The lateral shift of the vertical tire force
Fz
from the geometric contact point
is equivalent to a force applied in
Mx = Fz y
acting around a longitudinal axis in
P,
to the
and the tipping torque

(3.71)
Figure 3.15.
Note: Figure 3.15 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.
As long as the cambered tire has full contact to the road the lateral displacement
is
given by Eq. (3.40). Then, Eq. (3.71) reads as
Mx = Fz
b2 tan
.
12 4z cos
(3.72)
35
Vehicle Dynamics
en
ey
en
ey
P Q
y
Fz
Fz
Tx
Figure 3.15: Tipping torque at full contact
If the wheel load is approximated by its linearized static part

camber angles
|| 1
Mx = cN 4z
where the term
1
12
c N b2
Fz cN 4z
and small
are assumed, then, Eq. (3.72) simplies to
1
b2
=
c N b2 ,
12 4z
12
(3.73)
can be regarded as the tipping stiness of the tire.
en
ey
Q
P
Fz
Figure 3.16: 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
the contact area whereas the static contact point
3.16. In the following the static contact
may even be located outside
is still a real contact point, Figure
will be used as the contact point, because it
represents the contact area more precisely than the geometric contact point
P.
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, Figure 3.17. The resulting vertical
force
Fz
is applied in the center
torque around the
of the contact patch and hence, will not generate a
y -axis.
If the tire rotates tread particles will be stued into the front of the contact area which
causes a slight pressure increase, Figure 3.17. Now, the resulting vertical force is applied
in front of the contact point and generates the rolling resistance torque
ty = Fz xR sign() ,
36
(3.74)
rotating
en
ex non-rotating
xR
ex
en
C
Fz
Fz
Figure 3.17: Pressure distribution at a non-rotation and rotation tire
where
sign() assures
xR from C to
r0
distance
radius
that
ty
always acts against the wheel angular velocity
the working point of
fR =
Fz
The
usually is related to the unloaded tire
xR
.
r0
(3.75)
The dimensionless rolling resistance coecient slightly increases with the traveling velocity
of the vehicle
fR = fR (v) .
20 km/h < v < 200 km/h, the rolling
tires is in the range of 0.01 < fR < 0.02.
Under normal operating conditions,

coecient for typical passenger car
(3.76)
resistance
The rolling resistance hardly inuences the handling properties of a vehicle. But it plays
a major part in fuel consumption.
3.4 Friction Forces and Torques

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
the at bed runs with the velocity
vx .
and
The distance between the rim center and the at
bed is controlled to the loaded tire radius corresponding to the wheel load
Fz , Figure 3.18.
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 vx
and the bottom with the average transport velocity vt = rD . Depending on the velocity
dierence 4v = rD vx the tread particle is deected in longitudinal direction
A tread particle enters at the time
u = (rD vx ) t .
(3.77)
The time a particle spends in the contact patch can be calculated by
T =
L
,
rD ||
(3.78)
37
Vehicle Dynamics
rD
vx
rD
vx
L
u max
Figure 3.18: Tire on at bed test rig
where
denotes the contact length, and
T >0
is assured by
||.
The maximum deection occurs when the tread particle leaves the contact patch at the
time
t=T
umax = (rD vx ) T = (rD vx )
L
.
rD ||
(3.79)
The deected tread particle applies a force to the tire. In a rst approximation we get
Fxt = ctx u ,
where
ctx
(3.80)
represents the stiness of one tread particle in longitudinal direction.
On normal wheel loads more than one tread particle is in contact with the track, Figure
3.19a. The number
of the tread particles can be estimated by
p =
where
is the length of one particle and
a)
Figure 3.19:
(3.81)
denotes the distance between the particles.
b)
L
,
s+a
cxt * u
cut * u max
a) Particles, b) Force distribution,
Particles entering the contact patch are undeformed, whereas the ones leaving have the
maximum deection. According to Eq. (3.80), this results in a linear force distribution
38
versus the contact length, Figure 3.19b. The resulting force in longitudinal direction for
particles is given by
Fx =
1 t
p c umax .
2 x
(3.82)
Using the Eqs. (3.81) and (3.79) this results in
Fx =
1 L t
L
cx (rD vx )
.
2 s+a
rD ||
(3.83)
A rst approximation of the contact length L was calculated in Eq. (3.34). Approximating
1
F /c results in
the belt deformation by 4zB
2 z R
Fz
,
cR
L2 4 r 0
where
cR
(3.84)
denotes the radial tire stiness, and nonlinearities and dynamic parts in the tire
deformation were neglected. Now, Eq. (3.82) can be written as
Fx = 2
rD v x
r0 ctx
Fz
.
s + a cR
rD ||
(3.85)
The nondimensional relation between the sliding velocity of the tread particles in lonS
gitudinal direction vx = vx rD and the average transport velocity rD || form the
longitudinal slip
sx =
The longitudinal force
sx
Fx
(vx rD )
.
rD ||
(3.86)
is proportional to the wheel load
Fz
and the longitudinal slip
in this rst approximation
Fx = k Fz sx ,
where the constant
summarizes the tire properties
(3.87)
r0 , s, a, ctx
and
cR .
Eq. (3.87) 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, Figure 3.20.
small slip values

Fx = k * Fz* s x
L
moderate slip values

Fx = Fz * f ( s x )
high slip values

Fx = FG
L
t
Fx <= FH
adhesion
L
t
Fx = FG
Fx = FH
sliding
adhesion sliding
Figure 3.20: Longitudinal force distribution for dierent slip values

The resulting nonlinear function of the longitudinal force
Fx
versus the longitudinal slip

sx can be dened by the parameters initial inclination (driving stiness) dFx0 , location
M
S
S
sM
x and magnitude of the maximum Fx , start of full sliding sx and the sliding force Fx ,
Figure 3.21.
39
Vehicle Dynamics
Fx
M
adhesion
sliding
Fx
S
Fx dFx0
sSx
sM
x
sx
Figure 3.21: Typical longitudinal force characteristics
3.4.2 Lateral Slip, Lateral Force and Self Aligning Torque

Similar to the longitudinal slip
sx ,
given by Eq. (3.86), the lateral slip can be dened by
vyS
sy =
,
rD ||
(3.88)
where the sliding velocity in lateral direction is given by
vyS = vy
(3.89)
and the lateral component of the contact point velocity
vy
follows from Eq. (3.50).
As long as the tread particles stick to the road (small amounts of slip), an almost linear
distribution of the forces along the length
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, Figure 3.22.
Fy
large slip values

Fy = FG
sliding
Fy
adhesion
moderate slip values

Fy = Fz * f ( s y )
sliding
L
Fy n
adhesion
small slip values

Fy = k * Fz * s y
Figure 3.22: Lateral force distribution over contact patch
The nonlinear characteristics of the lateral force versus the lateral slip can be described
0
M
M
by the initial inclination (cornering stiness) dFy , the location sy and the magnitude Fy
40
of the maximum, the beginning of full sliding
sSy ,
and the magnitude
FyS
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
Fy
with the dynamic tire oset or pneumatic trail
as a lever
generates the self aligning torque
TS = n Fy .
The lateral force
sy .
Fy
(3.90)
as well as the dynamic tire oset are
functions of the lateral slip
Typical plots of these quantities are shown in Figure 3.23. Characteristic parameters
n/L
(n/L)0
Fy
M
adhesion
Fy
adhesion
adhesion/sliding
full sliding
adhesion/
sliding
s0y
full sliding
Fy dF0y
MS
sSy
sy
adhesion
adhesion/sliding
full sliding
sM
y
sSy
sy
s0y
sSy
sy
Figure 3.23: Typical plot of lateral force, tire oset and self aligning torque
M
0
of the lateral force graph are initial inclination (cornering stiness) dFy , location sy and
S
S
M
magnitude of the maximum Fy , begin of full sliding sy , and the sliding force Fy .
The dynamic tire oset has been normalized by the length of the contact patch L. The
0
S
initial value (n/L)0 as well as the slip values sy and sy suciently characterize the graph.
3.4.3 Wheel Load Influence

The resistance of a real tire against deformations has the eect 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
41
Vehicle Dynamics
Longitudinal force
Fz = 3.0 kN
Fx
Lateral force
Fz = 6.0 kN
Fz = 3.0 kN
Fy
Fz = 6.0 kN
dFx0 = 70 kN
dFx0 = 220 kN
dFy0 = 72 kN
dFy0 = 130 kN
sM
x = 0.160
sM
x = 0.120
sM
y = 0.180
sM
y = 0.200
FxM = 2.90 kN FxM = 5.60 kN FyM = 2.85 kN FyM = 5.40 kN

sSx = 0.500
sSx = 0.500
FxS = 2.65 kN FxS = 5.10 kN
sSy = 0.500
sSy = 0.700
FyS = 2.80 kN FyS = 5.30 kN
Table 3.3: Characteristic tire data with degressive wheel load inuence
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.
In practice, this leads to a degressive inuence of the wheel load on the characteristic
curves of longitudinal and lateral forces. In order to respect this fact in a tire model, the
N
N
characteristic data for two nominal wheel loads Fz and 2 Fz are given in Table 3.3.
M
M
0
0
From this data the initial inclinations dFx , dFy , the maximal forces Fx , Fx and the
M
S
sliding forces Fx , Fy for arbitrary wheel loads Fz are calculated by quadratic functions.
For the maximum longitudinal force it reads as

Fz
1 M
1 M
N
M
N Fz
N
N
M
= N 2 Fx (Fz ) 2 Fx (2Fz ) Fx (Fz ) 2 Fx (2Fz ) N .
Fz
Fz
Fy [kN]
Fx [kN]
FxM (Fz )
-2
-2
-4
-4
-6
-0.4
-0.2
0.2
0.4
-6
-20
-10
Figure 3.24: Longitudinal and lateral force characteristics:
M
sM
x , sy ,
10
20
Fz = 1.8, 3.2, 4.6, 5.4, 6.0 kN
sSx , sSy , at which full sliding

load Fz . For the location of the
and the slip values,
appears, are dened as linear functions of the wheel
42
[deg]
sx [-]
The location of the maxima
(3.91)
maximum longitudinal force this results in
sM
x (Fz )
N
sM
x (Fz )
N
sM
x (2Fz )

F
z
1 .
FzN
N
sM
x (Fz )
(3.92)
With the numeric values from Tab. 3.3 a slight shift of the maxima with an increasing
wheel load is also modeled, Figure 3.24.
3.4.4 Two-Dimensional Tire Characteristics

The longitudinal force as a function of the longitudinal slip
Fx = Fx (sx )
and the lateral
force depending on the lateral slip Fy = Fy (sy ) can be dened by their characteristic
0
0
M
M
parameters initial inclination dFx , dFy , location sx , sy and magnitude of the maximum
M
M
S
S
Fx , Fy as well as sliding limit sx , sy and sliding force FxS , FyS , Figure 3.25. During
general driving situations, e.g. acceleration or deceleration in curves, longitudinal sx and
sy
lateral slip
Fx
appear simultaneously.
Fx
Fx
Fy
0
dF x
sx
sSx
sM
x
Fy Fy
dF 0
FS
FM
Fy
F(s)
Fx
sy
dF y
sSy
sS
sy
sM
sM
y
sx
Figure 3.25: Generalized tire characteristics

The longitudinal slip
sx
and the lateral slip
sy
can vectorally be added to a generalized
slip
s
s =
where the slips
sx
and
sy
sx
sx
2

+
sy
sy
2
,
were normalized by appropriate weighting factors
(3.93)
sx
and
sy .
43
Vehicle Dynamics
Similar to the graphs of the longitudinal and lateral forces the graph of the generalized tire
0
M
M
S
S
force is dened by the characteristic parameters dF , s , F , s and F . The parameters
are calculated from the corresponding values of the longitudinal and lateral force
dF
sM =
FM =
sS =
FS =
q
2
(dFx0 sx cos )2 + dFy0 sy sin ,
s
2 M
2
sy
sM
x
cos +
sin ,
sx
sy
q
2
(FxM cos )2 + FyM sin ,
s
2 S
2
sy
sSx
cos +
sin ,
sx
sy
q
2
(FxS cos )2 + FyS sin ,
(3.94)
where the slip normalization have also to be considered at the initial inclination. The
angular functions
cos =
sx /
sx
s
and
sin =
sy /
sy
s
(3.95)
grant a smooth transition from the characteristic curve of longitudinal to the curve of
lateral forces in the range of = 0 to = 90 .
F = F (s) is now described in intervals by a broken rational function, a cubic

S
polynomial, and by the sliding force F

,
sM dF 0
= M ,
0 s sM ;
s
s
1 + + dF 0 M 2
F
F (s) =
(3.96)
s sM
M
M
S
2
M
S
(F
F
)
(3
2
)
,
=
,
s
<
s
s
;
sS sM
FS ,
s > sS .
The function
0
When dening the curve parameters, one just has to make sure that the condition dF
M
2 FsM is fullled, because otherwise the function has a turning point in the interval 0
s sM .
<
Now, the longitudinal and the lateral force follow from the according projections in longitudinal and lateral direction
Fx = F cos
and
Fy = F sin .
(3.97)
Hence, within TMeasy the one-dimensional characteristics are automatically converted to

a two-dimensional combination characteristics, Figure 3.26.
44
30
3
20
10
F [kN]
1
0
F [kN]
-1
-10
-2
-20
-3
-4
-2
0
F [kN]
-30
-20
0
F [kN]
20
|| = 1, 2, 4, 6, 10, 14
|sx | = 1, 2, 4, 6, 10, 15 %;
Figure 3.26: Two-dimensional tire characteristics at
Fz = 3.2 kN
Fz = 35 kN
3.4.5 Different Friction Coefficients

The tire characteristics are valid for one specic tire road combination only. Hence, different tire road combinations will demand for dierent model parameter.
Fy [N]
4000
L /0
3000
0.2
2000
1000
0.4
0.6
-1000
0.8
-2000
1.0
-3000
-4000
-0.5
sy [-]
0.5
Fz = 3.2 kN
Figure 3.27: Lateral force characteristics for dierent friction coecients

If only the coecient of friction is changed a simple but eective adaption of given model
data is possible. A reduced or changed friction coecient mainly inuences the maximum
force and the sliding force, whereas the initial inclination will remain unchanged. So, by
setting
sM
L M
L M
s , FM
F ,
0
0
sS
L S
L S
s , FS
F ,
0
0
(3.98)
45
Vehicle Dynamics
the essential tire model parameter which are valid for the friction coecient
justed to the new friction coecient
L .
are ad-
The result of this simple approach is shown in
Figure 3.27.
If the road model provides not only the roughness information
local friction coecient
[z, L ] = fR (x, y)
then, braking on
z = fR (x, y)
-split
but also the
maneuvers can easily
be simulated.
3.4.6 Self Aligning Torque

According to Eq. (3.90) the self aligning torque can be calculated via the dynamic tire
oset. The dynamic tire oset
can be normalized by the length
of the contact area,
nN = n/L. It mainly depends on the lateral slip sy . The normalized tire oset starts
at sy = 0 with an initial value (n/L)0 . It tends to zero, n/L 0 at large slip values,
sy sSy . Sometimes the dynamic tire oset overshoots to negative values before it reaches
S
0
zero again. This behavior can be modeled by introducing the parameter sy < sy , Figure
3.28.
n/L
n/L
(n/L)0
(n/L)0
s0y
sy
sSy
s0y
sy
Figure 3.28: Normalized tire oset with and without overshoot
In order to achieve a simple and smooth approximation of the normalized tire oset versus
0
the lateral slip, a linear and a cubic function are overlayed in the rst section sy sy

|sy |
(1w) (1s) + w 1 (32s) s2

|sy | s0y and s = 0
sy
2
0 S
n n
|sy | sy sy |sy |
=
(1w)
s0y < |sy | sSy
L
L 0
0
S s0
s
s
y
y
y
0
|sy | > sSy
where the factor
s0y
w= S
sy
(3.99)
(3.100)
0
weights the linear and the cubic function according to the values of the parameter sy
S
0
S
and sy . No overshoot occurs for sy = sy . Here, w = 1 and (1 w) = 0 will produce a
cubic transition from
sy = s0y .
46
n/L = (n/L)0
to
n/L = 0 with vanishing inclinations at sy = 0 and
The characteristic curve parameters, which are used for the description of the dynamic
tire oset, are at rst approximation not wheel load dependent. Similar to the description
of the characteristic curves of longitudinal and lateral force, here also the parameters for
single and double pay load are given. The calculation of the parameters of arbitrary wheel
loads is done similar to Eq. (3.92) by linear inter- or extrapolation.
Tz [Nm]
200
Tire oset parameter
150
100
Fz
50
Fz = 3.0 kN
Fz = 6.0 kN
(n/L)0 = 0.170 (n/L)0 = 0.190
0
-50
s0y = 0.200
s0y = 0.220
-100
sE
y = 0.420
sE
y = 0.400
-150
-200
-30
Figure 3.29: Self aligning torque:

The value of
(n/L)0
-20
-10
10
20
[deg]
30
Fz = 1.5, 3.0, 4.5, 6.0, 7.5 kN
can be estimated very well. At small values of lateral slip
sy 0
one
gets at rst approximation a triangular distribution of lateral forces over the contact area
length cf. Figure 3.22. The working point of the resulting force (dynamic tire oset) is
then given by
n(Fz 0, sy = 0) =
The value
1
L.
6
(3.101)
n = 16 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 oset.
The self aligning torque in Figure 3.29 has been calculated with the tire parameter from
Table 3.3. The degressive inuence of the wheel load on the self aligning torque can be
seen here as well.
With the parameters for the description of the tire oset it has been assumed that at
N
the payload Fz = Fz the related tire oset reaches the value of (n/L)0 = 0.17 1/6 at
sy = 0. The slip value s0y , at which the tire oset passes the x-axis, has been estimated.
Usually the value is somewhat higher than the position of the lateral force maximum.
S
With rising wheel load it moves to higher values. The values for sy are estimated too.
47
Vehicle Dynamics
3.4.7 Camber Influence

At a cambered tire, Figure 3.30, the angular velocity of the wheel
has a component
normal to the road
n = sin .
en
rim
centre
plane
eyR
(3.102)
Fy = Fy (s y ): Parameter
4000
3000
2000
1000
0
-1000
ex
ey
rD ||
y()
v()
-2000
-3000
-4000
-0.5
Figure 3.30: Cambered tire
Fy ()
at
0.5
Fz = 3.2 kN
and
= 0 , 2 , 4 , 6 , 8
Now, the tread particles in the contact patch possess a lateral velocity which depends on
their position
and is provided by
v () = n
L
, = sin ,
2 L/2
L/2 L/2 .
(3.103)
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 () .
(3.104)
The time derivative can be transformed to a space derivative
y () =
where
rD ||
(3.105)
denotes the average transport velocity. Now, Eq. (3.104) can be written as
d y ()
rD || = sin
d
48
d y () d
d y ()
=
rD ||
d dt
d
or
d y ()
sin L
=
,
d
rD || 2 L/2
(3.106)
where
L/2
was used to achieve dimensionless terms. Similar to the lateral slip
sy
which
is dened by Eq. (3.88) we can introduce a camber slip now
sin L
.
rD || 2
(3.107)
d y ()
= s
.
d
L/2
(3.108)
s =
Then, Eq. (3.106) simplies to
The shape of the lateral displacement prole is obtained by integration

2
1 L
+ C.
(3.109)
y = s
2 2 L/2

1
The boundary condition y = L = 0 can be used to determine the integration constant
2
C . One gets
1 L
C = s
.
(3.110)
2 2
Then, Eq. (3.109) reads as
1 L
y () = s
2 2
"

1
L/2
2 #
.
(3.111)
The lateral displacements of the tread particles caused by a camber slip are compared
now with the ones caused by pure lateral slip, Figure 3.31. At a tire with pure lateral
a) camber slip
y
b) lateral slip
yy()
y()
_
yy
_
y
-L/2
L/2
-L/2
L/2
Figure 3.31: Displacement proles of tread particles

slip each tread particle in the contact patch possesses the same lateral velocity which
dyy /d rD || = vy ,
y y was
dyy /d . Hence, the deection prole is linear, and
reads as yy = vy /(rD ||) = sy , where the denition in Eq. (3.88) was used to
introduce the lateral slip sy . Then, the average deection of the tread particles under
results in
where according to Eq. (3.105) the time derivative
transformed to the space derivative
pure lateral slip is given by
yy = sy
L
.
2
(3.112)
49
Vehicle Dynamics
The average deection of the tread particles under pure camber slip is obtained from
ZL/2 "
1 L 1
y = s
2 2 L

1
x
L/2
2 #
1
L
d = s .
3
2
(3.113)
L/2
A comparison of Eq. (3.112) with Eq. (3.113) shows, that by using
1
s
3
sy =
the lateral camber slip
(3.114)
can be converted to an equivalent lateral slip
sy .
In normal driving conditions, the camber angle and thus, the lateral camber slip are
limited to small values. So, the lateral camber force can be approximated by
Fy dFy0 sy .
If the global inclination
dFy = Fy /sy
(3.115)
is used instead of the initial inclination
dFy0 ,
one
gets the camber inuence on the lateral force as shown in Figure 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 dicult, 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

In particular during steering motions the angular velocity of the wheel
0W = 0R
+ eyR
has a component in direction of the track normal
(3.116)
en
n = eTn 0W 6= 0 .
(3.117)
Then, a very complicated deection prole of the tread particles in the contact patch
occurs. 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 tire's contact patch is approximated by a circle, Figure
3.32. By setting
1
R =
2
L B
+
2
2
1
(L + B)
4
length L and
the radius of the circle can be adjusted to the
(3.118)
the width
of the actual
contact patch. The integration over the whole circle area results in the bore torque
TB
50
1
=
A
Z
F r dA ,
A
(3.119)
ex
B
dr
F
r
L
ey
circular
approximation
normal shape of contact patch
Figure 3.32: Bore torque approximation
denotes the force transmitted by the patch element dA, and

2
circle. With dA = r d dr and A = R Eq. (3.119) reads as
where
TB
1
= 2
R
is the area of the
ZR Z2
F r rd dr
0
(3.120)
which immediately results in
TB
R
2
Z
ZR
1
2
F r r dr = 2
= 2
F r2 dr .
R
R
0
(3.121)
For small slip values the force transmitted in the patch element can be approximated by
F = F (s) dF 0 s
where
denotes the slip of the patch element, and
dF 0
(3.122)
is the initial inclination of the
generalized tire force characteristics. Similar to Eqs. (3.86 and (3.88) we dene
s =
where
r n
r n
rD ||
describes the sliding velocity in the patch element, and
(3.123)
rD
and
denote the
dynamic tire radius and the angular velocity of the wheel.

Now, Eq. (3.121) reads as
TB
2
= 2
R
ZR
dF 0
r n 2
r dr
rD ||
(3.124)
51
Vehicle Dynamics
which nally results in
TB
2
n
= 2 dF 0
R
rD ||
ZR
2
1
R n
n R4
0
r dr = 2 dF
=
R dF 0
R
rD || 4
2
rD ||
3
(3.125)
0
where
sB =
R n
rD ||
(3.126)
can be considered as bore slip. Via the initial inclination
dF 0
the bore torque
TB
takes
the actual tire properties into account.

The bore torque calculated by Eq. (3.125) is only a rst approximation. At large bore slips
S
the generalized tire force F is limited to the sliding force F . Then, Eq. (3.121) changes
to
TBmax
2
= 2
R
ZR
F S r2 dr =
2 S R3
2 S
=
F R.
F
R2
3
3
(3.127)
0
Due to the generalized sliding force
FS
the maximum bore torque
TBmax
depends on the
tire properties and the actual friction value. Now, the bore torque is given by
TB =
1
R n
R dF 0
2
rD ||
where according to Eq. (3.118) the circle radius

the width
|TB |
with
2 S
F R
3
(3.128)
can be replaced by the length
and
of the contact patch.
3.4.9 Typical Tire Characteristics

passenger car tire
truck tire
40
20
F [kN]
2
0
F [kN]
1.8 kN
3.2 kN
4.6 kN
5.4 kN
-2
-4
-6
-40
10 kN
20 kN
30 kN
40 kN
50 kN
-20
-40
-20
20
40
-40
-20
sx [%]
Figure 3.33: Longitudinal force:
Meas.,
sx [%]
20
40
TMeasy
The tire model TMeasy which is based on this approach, can be used for passenger car
tires as well as for truck tires. It approximates the characteristic curves
52
Fx = Fx (sx ),
passenger car
truck
40
Fy [kN]
F [kN]
20
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-2
-4
-40
100
1000
50
500
[Nm]
1500
[Nm]
150
-50
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-100
-150
-20
-10
[o]
10
20
10 kN
20 kN
30 kN
40 kN
-20
-6
Mz
-500
18.4 kN
36.8 kN
55.2 kN
-1000
-1500
-20
Figure 3.34: Lateral force and self aligning torque:
Fy = Fy ()
and
Mz = Mz ()
-10
[o]
Meas.,
quite well even for dierent wheel loads
10
20
TMeasy
Fz ,
Figures 3.33
and 3.34.
When experimental tire values are missing, the model parameters can be pragmatically
estimated by adjustment of the data of similar tire types. Furthermore, due to their physical signicance, the parameters can subsequently be improved by means of comparisons
between the simulation and vehicle testing results as far as they are available.
53
4 Suspension System
4.1 Purpose and Components
The automotive industry uses dierent kinds of wheel/axle suspension systems. Important
criteria 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 eect 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,
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.
54
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 eects 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 eect 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.
55
Vehicle Dynamics
4.2.2 Specific 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
characteristic. 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.
56
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
uR S
nk
drag li
wheel
and
wheel
body
rack
pinion
steering
box
Figure 4.4: Rack and pinion steering

wheel
uR = uR (S ) and then via

1 = 1 (uR ), 2 = 2 (uR ). Hence, the
are transformed by the steering box to the rack travel
the drag links transmitted to the wheel rotations
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
ste
e
ring
ever
l
ring
stee
lev
er 1
drag link 1
drag link 2
2
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
57
Vehicle Dynamics
suspension at the front axle. Here, the steering box can be placed outside of the axle
center.
Firstly, the rotations of the steering wheel
rotation of the steer levers
1 = 1 (L ), 2 = 2 (L ).
are transformed by the steering box to the
L = L (S ). The drag links transmit this rotation to the wheel
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
L
wheel
and
wheel
body
g
steerin
r
leve
steer box
O
(90o rotated)
steering link
2
drag link
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
wheel
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
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
wheel
is transferred by a coupling link to the right lever arm. Finally, the left and right wheels
are rotated via the drag links,
58
1 = 1 (A )
and
2 = 2 (A ).
steer
in
g leve
r
L
steering box
steering link
left
lever
arm
drag link
coupl.
link
2
wheel and
wheel body
Figure 4.7: Typical bus steering system
4.4 Standard Force Elements

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
FS
u
u
FS
Air spring
Torsion bar
Leaf spring
FS
FS
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.
59
Vehicle Dynamics
FS
FS
u
LF
FS0
L
c
L0
FS
Figure 4.9: Linear coil spring and general spring characteristics
A linear coil spring may be characterized by its free length
LF
and the spring stiness
c,
Fig. 4.9. The force acting on the spring is then given by

FS = c LF L ,
where
(4.1)
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
L0 < L F .
Now, Eq. (4.1) can be written as
FS = c LF (L0 u)
where
FS0
is the spring preload and

= c LF L0 + c u = FS0 + c u ,
(4.2)
describes the spring displacement measured from
the spring's conguration length.

In general the spring force
placement
FS
can be dened by a nonlinear function of the spring dis-
u
FS = FS (u) .
(4.3)
Now, arbitrary spring characteristics can be approximated by elementary functions, like

polynomials, 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,
FS = FS (u).
For detailed investigations sophisticated or
even dynamic spring models have to be used.
4.4.2 Damper
Dampers are basically oil pumps, Fig. 4.10. As the suspension travels up and down, the
hydraulic uid is forced by a piston through tiny holes, called orices. This slows down
the suspension movement.
60
Remote Gas Chamber
FD
Remote Oil Ch.

Remote orifice
Rebound Ch.
Piston
FD
Compression
Chamber
Piston orifice
v
Figure 4.10: Principle of a mono-tube damper
Today twin-tube and mono-tube dampers are used in vehicle suspension systems. Dynamic
damper models 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.
In standard vehicle dynamics applications simple characteristics
FD = FD (v)
are used to describe the damper force
FD
(4.4)
as a function of the damper velocity
v.
To
obtain this characteristics the damper is excited with a sinusoidal displacement signal
u = u0 sin 2f t. By varying the frequency in several steps from f = f0 to f = fE dierent

force displacement curves FD = FD (u) are obtained, Fig. 4.11. By taking the peak values
of the damper force at the displacement u = u0 which corresponds with the velocity
v = 2f u0 the characteristics FD = FD (v) is generated now. Here, the rebound cycle is
associated with negative damper velocities.
FD = FD(u)
FD = FD(v)
FD [N]
1000
0
Compression
f0
-1000
-2000
Rebound
-3000
fE
-4000
-0.06 -0.04 -0.02
0.02 0.04 0.06 -1.6
u [m]
-1.2
-0.8
-0.4
0.4
0.8
1.2
1.6
v [m/s]
Figure 4.11: Damper characteristics generated from measurements

Typical passenger car or truck dampers will have more resistance during its rebound cycle
then its compression cycle.
61
Vehicle Dynamics
4.4.3 Rubber Elements

Force elements made of natural rubber or urethane compounds are used in many locations
on the vehicle suspension system, Fig. 4.12. Those elements require no lubrication, isolate
minor vibration, reduce transmitted road shock, operate noise free, oer high load carrying
capabilities, and are very durable.
Topmount
Stop
Control arm
bushings
Subframe mounts
Figure 4.12: 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 movement. Thus, an additional resistance to
suspension movement 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 eects of the whole
axle.
It turns out, that those elastic elements can hardly be described by simple spring and
damper characteristics,
FS = FS (u) and FD = FD (v), because their stiness and damping
properties change with the frequency of the motion. Here, more sophisticated dynamic
models are needed.
62
4.5 Dynamic Force Elements

4.5.1 Testing and Evaluating Procedures
The eect 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
xe (t) = A sin(2 f t) ,
with dierent frequencies
f0 f fE
(4.5)
and amplitudes
Amin A Amax .
Starting at
t = 0,
the system will usually be in a steady state condition after several periods
where
T = 1/f
and
n = 2, 3, . . .
t nT ,
have to be chosen appropriately. Due to the nonlinear
F (t + T ) = F (T ), where T = 1/f , yet

F will be approximated
system behavior the system response is periodic,
not harmonic. That is why, the measured or calculated force

within one period
n T t (n + 1)T ,
by harmonic functions as good as possible
sin(2 f t) + cos(2 f t) .
F (t)
|
{z
}
|{z}
rst harmonic approximation
measured/
calculated
The coecients
1
2
(n+1)T
Z
and
(4.6)
can be calculated from the demand for a minimal overall error
2
sin(2 f t)+ cos(2 f t) F (t) dt
M inimum .
(4.7)
nT
The dierentiation of Eq. (4.7) with respect to
and
yields two linear equations as
necessary conditions
(n+1)T
Z
nT
(n+1)T
Z

sin(2 f t)+ cos(2 f t) F (t) sin(2 f t) dt = 0
(4.8)

sin(2 f t)+ cos(2 f t) F (t) cos(2 f t) dt = 0
nT
with the solutions
R
R
R
F sin dt cos2 dt F cos dt sin cos dt
R
R
R
=
sin2 dt cos2 dt 2 sin cos dt
,
R
R
R
R
F cos dt sin2 dt F sin dt sin cos dt
R
R
R
=
sin2 dt cos2 dt 2 sin cos dt
(4.9)
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. (4.9)
sin cos dt = 0 ;
sin2 dt =
R
T
T
;
cos2 dt =
2
2
(4.10)
63
Vehicle Dynamics
hold, and as solution
2
=
T
2
=
T
F sin dt ,
Z
F cos dt
(4.11)
remains. However, these are exactly the rst two coecients of a Fourier-"-Approximation.
The rst order harmonic approximation in Eq. (4.6) can now be written as
F (t) = F sin (2 f t + )
where amplitude
and phase angle
F =
(4.12)
are given by
2 + 2
tan =
and
A simple force element consisting of a linear spring with the stiness

with the constant
(4.13)
c and a linear damper
in parallel would respond with
F (t) = c xe + d x e = c A sin 2f t + d 2f A cos 2f t .
(4.14)
Here, amplitude and phase angle are given by
F = A
c2 + (2f d)2
Hence, the response of a pure spring,
tan =
and
c 6= 0
and
d=0
d
d 2f A
= 2f .
cA
c
(4.15)
F = A c and
d 6= 0 results in
is characterized by
tan = 0 or = 0, whereas a pure damper response with c = 0 and

F = 2f dA 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 stiness, dened by
cdyn =
F
A
(4.16)
is used to evaluate the stiness 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
xe (t) = A sin(2 f t) ,
where the period
T = 1/f
(4.17)
appears as pre-factor at dierentiation
x e (t) = A 2 f cos(2 f t) =
2
A cos(2 f t) .
T
(4.18)
A generalized sine-function can be constructed, now. Starting with
xe (t) = A sin(2 h(t)) ,
64
(4.19)
the time derivative results in
x e (t) = A 2 h(t)
cos(2 h(t)) .
In the following we demand that the function
time, i.e:
h(t)
=
where
p>0
and
q>0
generates periods fading linearly in
1
1
=
,
T (t)
pqt
(4.21)
are constants yet to determine. Eq. (4.21) yields
h(t) =
The initial condition
h(t)
(4.20)
h(t = 0) = 0
1
ln(p q t) + C .
q
(4.22)
xes the integration constant
C =
1
ln p .
q
(4.23)
With Eqs. (4.23) and (4.22) Eq. (4.19) results in a sine-like function
xe (t) = A sin
2
q
ln
p
,
pqt
(4.24)
which is characterized by linear fading periods.

The important zero values for determining the period duration lie at
1
p
ln
= 0, 1, 2,
q
p q tn
and
tn =
or
p
= en q ,
p q tn
mit
n = 0, 1, 2,
p
(1 en q ) , n = 0, 1, 2, .
q
(4.25)
(4.26)
The time dierence between two zero values yields the period
p
Tn = tn+1 tn = (1e(n+1) q 1+en q )
q
, n = 0, 1, 2, .
p n q
q
Tn =
e
(1 e )
q
For the rst
(n = 0)
and last
(n = N )
(4.27)
period one nds
p
(1 eq )
q
.
p
=
(1 eq ) eN q = T0 eN q
q
T0 =
TN
With the frequency range to investigate, given by the initial

the parameters
and the ratio
q/p
f0
(4.28)
and nal frequency
fE ,
can be calculated from Eq. (4.28)
1
fE
q =
ln
,
N
f0
n
hf iN o
q
E
= f0 1
,
p
f0
(4.29)
65
Vehicle Dynamics
with
xing the number of frequency intervals. The passing of the whole frequency range
then takes the time
1 e(N +1) q
.
q/p
tN +1 =
(4.30)
Hence, to test or simulate a force element in the frequency range from

with
N = 500
intervals will only take
728 s
or
0.1Hz to f = 100Hz
12min.
4.5.2 Simple Spring Damper Combination

Fig. 4.13 shows a simple dynamic force element where a linear spring with the stiness
and a linear damper with the damping constant
are arranged in series.
Figure 4.13: Spring and damper in series

The displacements of the force element and the spring itself are described by
and
s.
Then, the the forces acting in the spring and damper are given by
FS = c s
FD = FS
displacement s
The force balance

spring
and
FD = d (u s)
.
results in a linear rst order dierential equation for the
d (u s)
= cs
or
where the ratio between the damping coecient

constant,
(4.31)
d
d
s = s + u ,
c
c
d
and the spring stiness
(4.32)
acts as time
T = d/c. Hence, this force element will responds dynamically to any excitation.
The steady state response to a harmonic excitation
u(t) = u0 sin t
respectively
u = u0 cos t
(4.33)
can be calculated easily. The steady state response will be of the same type as the excitation. Inserting
s (t) = u0 (a sin t + b cos t)
(4.34)
into Eq. (4.32) results in
d
d
u0 (a cos t b sin t) = u0 (a sin t + b cos t) +
u0 cos t .
|
{z
}
|
{z
}
c
c | {z }
s
u
s
66
(4.35)
Collecting all sine and cosine terms we obtain two equations
d
u0 b = u0 a
c
d
d
u0 a = u0 b +
u0
c
c
and
(4.36)
which can be solved for the two unknown parameter
2
2 + (c/d)2
a =
and
b =
.
2
d + (c/d)2
(4.37)
Hence, the steady state force response reads as
FS = c s

c
sin t +
= c u0 2
cos t
d
+ (c/d)2
(4.38)
which can be transformed to
FS = FS sin (t + )
where the force magnitude
FS =
FS
and the phase angle
p
c u0
c u0
2 + (c/d)2 = p
2
2
2
+ (c/d)
+ (c/d)2
The dynamic stiness
cdyn = FS /u0
(4.39)
are given by
and
and the phase angle
= arctan
c/d
.
(4.40)
are plotted in Fig. 4.14 for
dierent damping values.
cdyn
400
4
300
[N/mm]
200
c = 400 N/mm
2
1
100
0
100
d
d1 = 1000 N/(m/s)
d2 = 2000 N/(m/s)
[o]
50
d1 = 3000 N/(m/s)
d2 = 4000 N/(m/s)
3
4
0
20
40
60
80
100
f [Hz]
Figure 4.14: Frequency response of a spring damper combination
With increasing frequency the spring damper combination changes from a pure damper
performance, cdyn 0 and 90 to a pure spring behavior, cdyn c and 0. The

frequency range, where the element provides stiness and damping is controlled by the
value for the damping constant
d.
67
Vehicle Dynamics
4.5.3 General Dynamic Force Model

To approximate the complex dynamic behavior of bushings and elastic mounts dierent
spring damper models can be combined. A general dynamic force model is constructed
by
parallel force elements, Fig. 4.15. The static load is carried by a single spring with
the stiness
c0
F0 = F0 (u).
or an arbitrary nonlinear force characteristics
u
FM
F1
d1
d2
FM
F2
dN
FM
FN
c0
s2
s1
c1
sN
cN
c2
Figure 4.15: Dynamic force model

Within each force element the spring acts in serial to parallel combination of a damper
and a dry friction element. Now, even hysteresis eects and the stress history of the force
element can be taken into account.
The forces acting in the spring and damper of force element
FSi = ci si
were
and
si
and
are given by
FDi = di (s i u)
,
(4.41)
describe the overall element and the spring displacement.
As long as the absolute value of the spring force FSi is lower than the maximum friction
M
force FF the damper friction combination will not move at all
u s i = 0
for
|FSi | FFM .
(4.42)
In all other cases the force balance
FSi = FDi FFM
(4.43)
holds. Using Eq. 4.41 the force balance results in
di (s i u)
= FSi FFM
(4.44)
which can be combined with Eq. 4.42 to
F + FFM
Si
di s i =
di u
F F M
Si
68
FSi < FFM

for
FFM FSi +FFM

+FFM < FSi
(4.45)
where according to Eq. 4.41 the spring force is given by
FSi = ci si .
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 hydromounts are used. The dynamic nonlinear behavior of these components guarantees a
good acoustic decoupling but simultaneously provides sucient damping.
xe
main spring
chamber 1
membrane
cF
cT
__
2
ring channel
chamber 2
uF
c__
T
2
MF
dF
__
2
dF
__
2
Figure 4.16: Hydro-mount

Fig. 4.16 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 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 eect of a hydro-mount is combined from the elasticity of the main spring and
the volume change in chamber 1.
With
uF
labeling the displacement of the generalized uid mass
FH = cT xe + FF (xe uF )
MF ,
(4.46)
holds, where the force eect of the main spring has been approximated by a linear spring
with the constant
cT .
69
Vehicle Dynamics
With
MF R
as the actual mass in the ring channel and the cross sections
A K , AR
of
chamber and ring channel the generalized uid mass is given by
MF =
A 2
K
AR
MF R .
(4.47)
The uid in chamber 1 is not being compressed, unless the membrane can evade no longer.
With the uid stiness
FF (xe uF ) =
cF and the membrane clearance sF , one gets

c
(x
u
)
+
s
(xe uF ) < sF
F
e
F
F
0

c (x u ) s
F
e
F
F
The hard transition from clearance

tion with
FF 6= 0
for
|xe uf |
sF
(4.48)
(xe uf ) > +sF
FF = 0 and uid compression resp. chamber deforma-
is not realistic and leads to problems, even with the numeric solution.
Therefore, the function (4.48) is smoothed by a parabola in the range
|xe uf | 2 sF .
The motions of the uid mass cause friction losses in the ring channel, which are as a rst
approximation proportional to the speed,
FD = dF u F .
(4.49)
Then, the equation of motion for the uid mass reads as
MF uF = FF FD .
(4.50)
The membrane clearance makes Eq. (4.50) nonlinear and only solvable by numerical integration. The nonlinearity also aects the overall force in the hydro-mount, Eq. (4.46).
The dynamic stiness and the dissipation angle of a hydro-mount are displayed in Fig. 4.17
versus the frequency.
The simulation is based on the following system parameters
mF
25 kg
generalized uid mass
cT
125 000 N/m
stiness of main spring
dF
750 N/(m/s)
damping constant
cF
100 000 N/m
uid stiness
sF
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.
70
400
Dynamic Stiffness [N/m] at Excitation Amplitudes A = 2.5/0.5/0.1 mm
300
200
100
0
60
Dissipation Angle [deg] at Excitation Amplitudes A = 2.5/0.5/0.1 mm
50
40
30
20
10
0
10
Excitation Frequency [Hz]
10
Figure 4.17: Dynamic stiness [N/mm] and dissipation angle [deg] for a hydro-mount
71
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 sucient safety.
The stress of the load can be judged fairly well by maximal or integral values of the body
accelerations.
The wheel load
Fz
is linked to the longitudinal
of friction. The digressive inuence of

at the increase of
Fx
and
Fy
Fz
on
Fx
Fx
and
and lateral force
Fy
Fy
by the coecient
as well as non-stationary processes
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 dierent approaches of describing the human sense of vibrations by dierent metrics can be found.
Transferred to vehicle vertical dynamics, the driver primarily registers the amplitudes and
accelerations of the body vibrations. These values are thus used as objective criteria in
practice.
5.2 Modelling Aspects

5.2.1 Full Vehicle Model
For detailed investigations of ride comfort and ride safety sophisticated road and vehicle
models are needed. 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.
72
Ford

Time =
0.000000
ZZ
Y
Y
X
X
Figure 5.1: Full vehicle model
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
5.2.2 Twodimensional Models
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 per-
a2
a1
M2
zC1
M1
M*
zB
zA2
m2
M,
zR(s-a2)
zA1
zR(s+a1)
zC2
pitch
m1
hub
C
yB
xB
zR(s)
Figure 5.2: Vehicle model for basic comfort and safety analysis
forms hub and pitch motions. Here, the chassis is considered as one rigid body. Then,
mass and inertia properties 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
zA1
and
zA2
zC1
zC2 . The coordinates

axle. The function zR (s)
and in the rear
describe the vertical motions of the front and rear
provides road irregularities in the space domain, where
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
zR (s a2 )
gravity C .
zR (s + a1 )
locate the position of the chassis center of
and
respectively, where
a1
and
a2
73
Vehicle Dynamics
The point masses must add up to the chassis mass
M1 + M + M2 = M
(5.1)
and they have to provide the same inertia around an axis located in the chassis center
and pointing into the lateral direction
a21 M1 + a22 M2 = .
(5.2)
The correct location of the center of gravity is assured by
a1 M 1 = a2 M 2 .
(5.3)
Now, Eqs. (5.2) and (5.3) yield the main masses
M1 =
a1 (a1 +a2 )
and
M2 =
,
a2 (a1 +a2 )
(5.4)
and the coupling mass
M = M
1
M a1 a2

(5.5)
follows from Eq. (5.1).
= M a1 a2 then,
the coupling mass vanishes M = 0, and the vehicle can be represented by two uncoupled
If the mass and the inertia properties of a real vehicle happen to result in
two mass systems describing the vertical motion of the axle and the hub motion of the
chassis mass on top of each axle.
mid
full
sports
size
size
utility
car
car
vehicle
m1 [kg]
80
100
125
120
600
m2 [kg]
80
100
125
180
1100
a1 [m]
a2 [m]
1.10
1.40
1.40
1.40
1.45
1.38
1.90
1.40
2.90
1.90
M [kg]
1100
1400
1950
3200
14300
[kg m2 ]
1500
2350
3750
5800
50000
M1
M [kg]
M2
545
126
429
600
200
600
914
76
960
925
1020
1255
3592
5225
5483
vehicles
properties
front axle
mass
rear axle
mass
center
of
gravity
chassis
mass
chassis
inertia
lumped
mass
model
commercial
heavy
vehicle
truck
Table 5.1: Mass and inertia properties of dierent vehicles
74
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 < M1
and
M < M2 .
Here, the two
mass model or the quarter car model represent a quite good approximation to the lumped
mass model. For commercial vehicles and trucks, where the coupling mass has the same
magnitude as the corresponding chassis masses, the quarter car model serves for basic
studies only.
5.2.3 Simple Models

At most vehicles, c.f. Table 5.1, the axle mass is much smaller than the corresponding
chassis mass,
mi Mi , i = 1, 2.
Hence, for a rst basic study axle and chassis motions
can be investigated independently. The quarter car model is now further simplied to two
single mass models, Fig. 5.3.
``
6
zC
M
`
cS ``
`
`
`
cS `
`
dS
dS
6
zW
`
c
6
zR
``
`` c
`` T
c
6
zR
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 zC + dS zC + cS zC = dS zR + cS zR
(5.6)
m zW + dS zW + (cS + cT ) zW = cT zR ,
(5.7)
and
where
zC
and
zW
label the vertical motions of the corresponding chassis mass and the
wheel mass with respect to the steady state position. The constants
cS , dS
describe the
suspension stiness and damping. The dynamic wheel load is calculated by
FTD = cT (zR zW )
(5.8)
75
Vehicle Dynamics
where
cT
is the vertical or radial stiness of the tire and
zR
denotes the road irregularities.
In this simple approach the damping eects in the tire are not taken into account.
5.3 Basic Tuning

5.3.1 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
zR = 0
zR = 0.
and
The remaining homogeneous second order dierential
equations can be written in a more general form as
z + 2 0 z + 02 z = 0 ,
(5.9)
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:
Wheel:
z zC ,
C =
z zW ,
W =
d
S
,
2 cS M
dS
2
(cS +cT )m
2
02 0C
=
2
02 0W
=
cS
;
M
cS +cT
.
m
(5.10)
The solution of Eq. (5.9) is of the type
z(t) = z0 et ,
where
z0
and
(5.11)
are constants. Inserting Eq. (5.11) into Eq. (5.9) results in
(2 + 2 0 + 02 ) z0 et = 0 .
Non-trivial solutions
z0 6= 0
(5.12)
are possible, if
2 + 2 0 + 02 = 0
will hold. The roots of the characteristic equation (5.13) depend on the value of
<1 :
1 :
p
1,2 = 0 pm i 0 1 2 ,

p
2
1,2 = 0 1 .
(5.13)
(5.14)
Figure 5.4 shows the root locus of the eigenvalues for dierent values of the viscous
damping rate
76
=0
=0.5
=0.2
=0.7
-3
=1.25
-2.5
-2
-1
-1.5
-0.5
=0.9
=0.2
=0.5
1 the eigenvalues 1,2
Re()/0
-0.5
=0.7
For
1.0
=1.25 =1.5
=1
Figure 5.4: Eigenvalues
Im()/0
0.5
=0.9
=1.5
and
=0
-1.0
for dierent values of
are both real and negative. Hence, Eq. (5.11) will produce a
< 1 holds, the eigenvalues 1,2 will become

complex conjugate of 1 . Now, the solution can be written as

p
0 t
2
z(t) = A e
sin 0 1 t ,
exponentially decaying solution. If

where
is the
complex,
(5.15)
A and are constants which have to be adjusted to given initial conditions z(0) = z0
and z(0)
= z0 . The real part Re (1,2 ) =

p 0 is negative and determines the decay of
1 2 part denes the actual frequency of
the solution. The imaginary Im (1,2 ) = 0
where
the vibration. The actual frequency
= 0
tends to zero,
0,
1 2
if the viscous damping ratio will approach the value one,
(5.16)
1.
In a more general way the relative damping may be judged by the ratio
Re(1,2 )
.
| 1,2 |
D =
(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 | =
r
Re(1,2 )2 + Im(1,2 )2 =
p
2
( 0 )2 + 0 1 2
= 0 ,
(5.19)
and hence, Eq. (5.17) results in
D =
For
+ 0
= .
0
(5.20)
1 the eigenvalues become real and negative. Then, Eq. (5.17) will always produce
D = 1. In this case the viscous damping rate is more
the relative damping value

sensitive.
77
Vehicle Dynamics
5.3.2 Spring Rates

5.3.2.1 Minimum Spring Rates
The suspension spring is loaded with the corresponding vehicle weight. At linear spring
characteristics the steady state spring deection is calculated from
u0 =
Mg
.
cS
(5.21)
At a conventional suspension without niveau regulation a load variation

leads to changed spring deections
In analogy to (5.21) the additional
4M g
.
cS
(5.22)
deection follows from
4u =
If for the maximum load variation
M M + 4M
u0 u0 + 4u.
4M
the additional spring deection is limited to
4u
the suspension spring rate can be estimated by a lower bound
cS
4M g
.
4u
(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
wheel by
500 kg are going to the

4MF = 150 kg/2 = 75 kg and each rear
of the permissible load of
front axle. Then, each front wheel is loaded by
4MR = (500 150) kg/2 = 175 kg .

u umax . At standard passenger cars it
umax 0.10 m. By setting 4u = umax /2 we demand
The maximum wheel travel is limited,

range of
umax 0.8 m
to
is in the
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
cmin
= ( 75 kg 9.81 m/s2 )/(0.08/2) m = 18400 N/m .
S
(5.24)
4MR = 175 kg . Assuming

umax 0.10 m the minimum
The maximum load over one rear wheel was estimated here by
that the suspension travel at the rear axle is slightly larger,
spring rate at the rear axle can be estimated by
cmin
= ( 175 kg 9.81 m/s2 )/(0.10/2) m = 34300 N/m ,
S
(5.25)
which is nearly two times the minimum value of the spring rate at the front axle. In order
to reduce this dierence we will choose a spring rate of
cS = 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
M1 = M2 = 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
78
the same chassis mass, to distribute
equally to the front and the rear. Then, the
corresponding cassis mass of a quarter car model is given here by

M = M1 + 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
2
is then given by 0C = cS /M . Hence, for a spring rate of cS = 20000 N/m the undamped
natural frequency of the unloaded car amounts to
f0C =
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,
f0C =
p
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 = M2 + M /2 /2 = (600 kg + 200/2 kg)/2 = 350 kg .
(5.29)
Now the undamped natural frequencies for the unloaded
0
f0C
=
p
34300 N/m/350 kg/(2 ) = 1.6 Hz
(5.30)
and the loaded car
L
f0C
=
34300 N/m/(350 + 175) kg/(2 ) = 1.3 Hz
(5.31)
are larger and dier more.
5.3.2.2 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
u0
can be modeled by the nonlinear
function
FS = FS0 + c0 u
where
FS0
is the spring preload,
cS

1+k
u
4u
2 !
,
describes the spring rate at
(5.32)
u = 0,
k > 0 characat u = 4u the
and
terizes the intensity of the nonlinearity. The linear characteristic provides

FSlin (4u) = FS0 + cS 4u. To achieve the same value with the nonlinear spring
value
FS0 + c0 4u (1 + k) = FS0 + cS 4u
must hold, where
cS
or
c0 (1 + k) = cS
(5.33)
describes the spring rate of the corresponding linear characteristics.
The local spring rate is determined by the derivative
dFS
= c0
du

1 + 3k
u
4u
2 !
.
(5.34)
79
Vehicle Dynamics
FS
dFS
du
8000
u=u
FS [N]
6000
M g
63 kN/m
44 kN/m
cS
FS0
dFS
du
4000
29 kN/m
u=0
20 kN/m
2000
0.05
u [m]
0.1
Figure 5.5: Principle and realizations of nonlinear spring characteristics
Hence, the spring rate for the loaded car at
u = 4u
is given by
cL = c0 (1 + 3 k) .
The intensity of the nonlinearity
(5.35)
can be xed, for instance, by choosing an appropriate
spring rate for the unloaded vehicle. With
c0 = 20000 N/m
the spring rates on the front
and rear axle will be the same for the unloaded vehicle. With
yields
k =
cS = 34300 N/m
Eq. (5.33)
34300
cS
1 = 0.715 .
1 =
c0
20000
(5.36)
The solid line in Fig. 5.5 shows the resulting nonlinear spring characteristics which is
characterized by the spring rates
3 0.715) = 62 900 N/m
c0 = 20 000 N/m
and
cL = c0 (1 + 3k) = 20 000 (1 +
for the unloaded and the loaded vehicle. Again, the undamped
natural frequencies
s
0
f0C
=
20000 N/m 1
= 1.20 Hz
350 kg
2
s
or
L
f0C
=
92000 N/m 1
= 1.74 Hz
(350+175) kg 2
(5.37)
for the unloaded and the loaded vehicle dier quite a lot.
The unloaded and the loaded vehicle have the same undamped natural frequencies if
c0
cL
=
M
M + 4M
or
cL
M + 4M
=
c0
M
(5.38)
will hold. Combing this relationship with Eq. (5.35) one obtains
M
1 + 3k =
M + 4M
80
or
1
k =
3

M + 4M
1 4M
1 =
.
M
3 M
(5.39)
Hence, for the quarter car model with

sity of the nonlinear spring amounts to
M = 350 kg and 4M = 175

k = 1/3 175/350 = 0.1667.
the intenThis value
cS = 34300 N/m will produce the dotted line in Fig. 5.5. The spring rates
c0 = cS /(1 + k) = 34 300 N/m / (1 + 0.1667) = 29 400 N/m for the unloaded and
cL = c0 (1 + 3k) = 29 400 N/m (1 + 3 0.1667) = 44 100 N/m for the loaded vehicle fol-
and
low from Eqs. (5.34) and (5.35). Now, the undamped natural frequency for the unloaded
p
p
0
0
= cL /(M + 4M ) = 1.46 Hz are
f0C
= c0 /M = 1.46 Hz and the loaded vehicle f0C
in deed the same.
5.3.3 Influence 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
and wheel displacement zW (t) as well as the chassis acceleration z
C
0
D
and the wheel load FT = FT + FT for dierent damping rates C and W . The dynamic
0
wheel load follows from Eq. (5.8), and the static wheel load is given by FT = (M + m) g ,
of the chassis
where
zC (t)
labels the constant of gravity.
To achieve the same damping rates for the chassis and the wheel model dierent values
for the damping parameter
dS
were needed.
With increased damping the overshoot eect in the time history of the chassis displacement and the wheel load becomes smaller and smaller till it vanishes completely at
and
W = 1.
The viscous damping rate
C = 1
=1
5.3.4 Optimal Damping

5.3.4.1 Avoiding Overshoots
If avoiding overshoot eects is the design goal then,
ratio. For
=1
=1
will be the optimal damping
the eigenvalues of the single mass oscillator change from complex to real.
Thus, producing a non oscillating solution without any sine and cosine terms.
C = 1 and W = 1 results in the optimal damping parameter

p
p
C =1
W =1

dopt
=
2
cS M , and dopt
(cS +cT )m .
(5.40)
S Comfort
S Safety = 2
According to Eq. (5.10)
So, the damping values
C =1
N

dopt
S Comfort = 5292
m/s
and
W =1
N

dopt
S Safety = 6928
m/s
will avoid an overshoot eect in the time history of the chassis displacement
in the time history of the wheel load
FT (t).
(5.41)
zC (t) or in the
Usually, as it is here, the damping values for
optimal comfort and optimal ride safety will be dierent. Hence, a simple linear damper
can either avoid overshoots in the chassis motions or in the wheel loads.
81
Vehicle Dynamics
chassis model
350 kg
dS
20000 N/m
wheel model
C [ - ]
dS [Ns/m]
W [ - ]
dS [Ns/m]
0.25
0.50
0.75
1.00
1.25
1323
2646
3969
5292
6614
0.25
0.50
0.75
1.00
1.25
1732
3464
5196
6928
8660
displacement [mm]
200
15
100
10
50
-50
-5
-100
50 kg
220000 N/m
displacement [mm]
20
150
20000 N/m
dS
-10
0
0.5
t [s]
0.05
0.1
t [s] 0.15
0.1
t [s]
wheel load [N]
acceleration [g]
1.5
6000
5000
0.5
4000
0
3000
2000
-0.5
1000
-1
0
0
0.5
t [s]
1.5
0.05
0.15
Figure 5.6: Time response of simple vehicle models to initial disturbances
The overshot in the time history of the chassis accelerations
zC (t)
will only vanish for
which surely is not a desirable conguration, because then, it takes a very long
time till the initial chassis displacement has fully disappeared.
5.3.4.2 Fast Approach to Steady State

Instead of avoiding overshoot eects 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
and
82
z(t
= 0) = 0.
z(t = 0) = z0
z0
zS
tE
z(t)
Figure 5.7: Evaluating a damped vibration
Counting the dierences of the system response
z(t) from the steady state value zS = 0 as
errors allows to judge the attenuation. If the overall quadratic error becomes a minimum
t=t
Z E
=
z(t)2 dt M in ,
(5.42)
t=0
the system approaches the steady state position as fast as possible. In theory
tE
holds, for practical applications a nite
tE
have to be chosen appropriately.
To judge ride comfort and ride safety the hub motion of the chassis zC , its acceleration
zC and the variations of the dynamic wheel load FTD can be used. In the absence of road
D
irregularities zR = 0 the dynamic wheel load from Eq. (5.8) simplies to FT = cT zW .
Hence, the demands
2C
t=t
Z E
2
g1 zC
g2 zC
2 i
dt M in
(5.43)
t=0
and
t=t
Z E
2S =
cT zW
2
dt M in
(5.44)
t=0
will guarantee optimal ride comfort and optimal ride safety. By the factors
g1
and
g2
the
acceleration and the hub motion can be weighted dierently.

The equation of motion for the chassis model can be resolved for the acceleration

2
zC + 2C zC ,
zC = 0C
where, the system parameter
C 0C = dS /(2M )
M , dS
and
cS
(5.45)
were substituted by the damping rate
and by the undamped natural frequency
0C = cS /M .
C =
Then, the
problem in Eq. (5.43) can be written as
2C =
t=t
Z E
t=0
t=t
Z E
=
t=0
2
g12 0C
zC + 2C zC
zC zC
g12
2
2 2
(0C
)
g12
2
0C
i
+ g22 zC2 dt
g22
2C
g12
g12
2
0C
2C
2
(2C )
zC
zC
(5.46)
M in ,
83
Vehicle Dynamics
where

xTC = zC zC
is the state vector of the chassis model. In a similar way Eq. (5.44)
can be transformed to
2S
t=t
Z E
c2T
2
zW
t=t
Z E
dt =
zW zW
c2T 0
0 0
#"
t=0
t=0
where
"
xTW = zW zW
zW
zW
#
M in ,
(5.47)
denotes the state vector of the wheel model.
The problems given in Eqs. (5.46) and (5.47) are called disturbance-reaction problems.
They can be written in a more general form
t=t
Z E
xT (t) Q x(t) dt M in
(5.48)
t=0
where
x(t)
denotes the state vector and
Q = QT
is a symmetric weighting matrix. For
single mass oscillators described by Eq. (5.9) the state equation reads as

0
1
z
z
=
.
2
0 2
z
z
|
{z
} | {z }
| {z }
x
x
A

(5.49)
the time response of the system exposed to the initial disturbance x(t = 0) =
x0 vanishes x(t ) = 0, and the integral in Eq.(5.48) can be solved by
For tE
t=t
Z E
xT (t) Q x(t) dt = xT0 R x0 ,
(5.50)
t=0
where the symmetric matrix
R = RT
is given by the Ljapunov equation
AT R + R A + Q = 0 .
(5.51)
For the single mass oscillator the Ljapunov equation
0 02
1 2

R11 R12
R12 R22

+
R11 R12
R12 R22

0
1
02 2

+
Q11 Q12
Q12 Q22

.
(5.52)
results in 3 linear equations
02 R12 02 R12 + Q11 = 0

02 R22 + R11 2 R12 + Q12 = 0
R12 2 R22 + R12 2 R22 + Q22 = 0
which easily can be solved for the elements of

R11 =
84
1
+
2
0 4

Q11 Q12 +
02
Q22 ,
4
(5.53)
R12 =
Q11
,
202
R22 =
Q11
Q22
+
.
2
4 0
4
(5.54)
For the initial disturbance
t=t
Z E
x0 = [ z0 0 ]T
x (t) Q x(t) dt =
z02
R11 =
Eq. (5.50) nally results in
z02

t=0
+
2
0 4

02
Q11 Q12 +
Q22 .
4
(5.55)
Now, the integral in Eq. (5.46) evaluating the ride comfort is solved by
2C

2
2
C
1
0C
2
2
2
2
2 2
2
=
+
g1 0C + g2 g1 0C 2 C +
g (2C )
2
0C
4C
4C 1
#
"

2 !
2
g
g
2
2
0C
2
2
g12 +
+
C 0C .
= z0C
0C
2
2
4C
0C
0C
2
z0C

where the abbreviation

By setting
g1 = 1
and
was nally replaced by
g2 = 0
(5.56)
C 0C .
the time history of the chassis acceleration
zC
is weighted
only. Eq. (5.56) then simplies to

2
2 0C
2C z = z0C
0C
C
4C
which will become a minimum for
0C 0
(5.57)
C .
or
achieved by a soft suspension spring
cS 0
C
0C 0 is
As mentioned before,
surely is not a desirable conguration. A low undamped natural frequency

or a large chassis mass
M .
However, a
large chassis mass is uneconomic and the suspension stiness 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
g1 = 0 and g2 = 1 weights the time history of the chassis displacement
only. Then, Eq. (5.56) results in
2C z
C
z2
= 0C
0C
1
+ C
4C
0C or

2
d 2C |zC
z0C
1
=
+1 = 0.
d C
0C 4C2
(5.58)
which will become a minimum for
A high undamped natural frequency
(5.59)
0C
contradicts the demand for rapidly van1

ishing accelerations. The viscous damping ratio C =
solves Eq. (5.59) and minimizes
2
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
accelerations 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
cS zC
are added. Hence,
g1 = M
and
g1 = 1
g2 = cS
and
M zC
and
or
g2 =
cS
2
= 0C
M
(5.60)
85
Vehicle Dynamics
provide system-tted weighting factors. Now, Eq. (5.56) reads as
2C
2
z0C
2
0C
0C
+ C 0C
2C
Again, a good ride comfort will be achieved by
0C 0.
(5.61)
For nite undamped natural
frequencies Eq. (5.61) becomes a minimum, if the viscous damping rate
d 2C |zC
2
2
0C
= z0C
d C
Hence, a viscous damping rate of
C =
0C
+ 0C
2C2
will satisfy

= 0.
(5.62)
1
2
2
(5.63)
or a damping parameter of
1
opt C = 2 2
dS Comfort
2 cS M ,
(5.64)
will provide optimal comfort by minimizing the merit function in Eq. (5.61).
For the passenger car with
parameter will amount to
which is
70%
M = 350 kg
and
cS = 20 000 N/m
the optimal damping
C = 12 2
N

dopt
S Comfort = 3742
m/s
(5.65)
of the value needed to avoid overshot eects in the chassis displacements.
The integral in Eq. (5.47) evaluating the ride safety is solved by
2S
z2
= 0W
0W

W
1
+
4W
c2T
(5.66)
where the model parameter m, cS , dS and cT where replaced by the undamped natural
2
frequency 0W = (cS + cT )/m and by the damping ratio W = W 0W = dS /(2m).
2
A soft tire cT 0 make the safety criteria Eq. (5.66) small S 0 and thus, reduces the
dynamic wheel load variations. However, the tire spring stiness 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
cS
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
if
cT
and
0W
the merit function in Eq. (5.66) will become a minimum
2
2S
z0W
=
W
0W
1
1+ 2
4W
c2T = 0
(5.67)
will hold. Hence, a viscous damping rate of
W =
86
1
2
(5.68)
or the damping parameter
p

dopt
(cS + cT ) m
S Safety =
(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
cS = 20 000 N/m
the optimal damping
parameter will now amount to
W = 12
N

dopt
S Safety = 3464
m/s
which is
50%
(5.70)
of the value needed to avoid overshot eects in the wheel loads.
5.4 Sky Hook Damper

5.4.1 Modelling Aspects
In standard vehicle suspension systems the damper is mounted between the wheel and
the body. Hence, the damper aects body and wheel/axle motions simultaneously.
sky
zC
zC
dB
dW
cS
zW
zR
cS
dS
FD
zW
m
cT
zR
a) Standard Damper
m
cT
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 5.2.3 must be
combined to a more enhanced model, Fig. 5.8a.
Assuming a linear characteristics the suspension damper force is given by
FD = dS (zW zC ) ,
where
dS denotes the damping constant, and zC , zW
(5.71)
are the time derivatives of the absolute
vertical body and wheel displacements.
87
Vehicle Dynamics
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
FD = dW zW dC zC ,
dS
where instead of the single damping constant
(5.72)
now two design parameter
dW
and
dC
are available.
The equations of motion for the quarter car model are given by
M zC = FS + FD M g ,
(5.73)
m zW = FT FS FD m g ,
M, m
ments, and g
where
are the sprung and unsprung mass,
zC , zW
denote their vertical displace-
is the constant of gravity.
The suspension spring force is modeled by
FS = FS0 + cS (zW zC ) ,
where
FS0 = mC g
is the spring preload, and
cS
(5.74)
the spring stiness.
Finally, the vertical tire force is given by
FT = FT0 + cT (zR zW ) ,
where
FT0 = (M + m) g
cS
is the tire preload,
the road roughness. The condition
FT 0
(5.75)
the vertical tire stiness, and
zR
describes
takes the tire lift o into account.
5.4.2 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
zC
zW
z
C
zW
| {z
x
= 0
cS
cS
M
cS
cS +cT
m
m
{z
A
dMC
dC
m
zC
h i

1
zW + 0 zR ,
0
dW
z
| {z }
M C
u
dW
cT
m
zW
m
} | {z }
| {z
}
x
B
(5.76)
M g , mg were compensated by the preloads FS0 , FT0 , the term B u

excitation, x denotes the state vector, and A is the state matrix. In this
where the weight forces

describes the
linear approach the tire lift o 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
88
Damping ratio = D
Frequencies [Hz]
12
10
Wheel
350 kg
0.8
0.7
0.6
dS 0.5
20000 N/m
50 kg
220000 N/m
Chassis
0.4
0.2
3880
3220
1000
2000
3000
4000
5000
1000
2000
3000
4000
5000
dS [N/(m/s)]
dS [N/(m/s)]
Figure 5.9: Quarter car model with standard damper
= Im(), and according to Eq. (vdyn-eq:

= D = Re()/ ||. evaluates the damping ratio.
by the imaginary parts,

ratio lambda)
By setting
dC = dS
and
dW = dS
relative damping
Eq. (5.76) represents a quarter car model with the
standard damper described by Eq. (5.71). Fig. 5.9 shows the eigenfrequencies
and the damping ratios
= D
f = /(2)
dS .
for dierent values of the damping parameter
C = 21 2 0.7 for the chassis motion

parameter dS = 3880 N/(m/s), and the damping
Optimal ride comfort with a damping ratio of

could be achieved with the damping
parameter
W = 0.5
dS = 3220 N/(m/s)
would provide for the wheel motion a damping ratio of
which correspond to minimal wheel load variations. This damping parameter
are very close to the values
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
rameter. Their inuence to the eigenfrequencies
dW
and
dS
two design pa-
and the damping ratios
is shown in
Fig. 5.10.
The the sky-hook damping parameter
dC , dW
the damping ratios. The chassis damping ratio

damping ratio
mainly depends on
dW .
have a nearly independent inuence on
mainly depends on
dC ,
and the wheel
Hence, the damping of the chassis and the
wheel motion can be adjusted to nearly each design goal. Here, a sky-hook damper with
dC = 3900 N/(m/s) and dW = 3200 N/(m/s) would generate the damping ratios dC = 0.7
and dW = 0.5 hence, combining ride comfort and ride safety within one damper layout.
89
Vehicle Dynamics
Damping ratios C, W
Frequencies [Hz]
12
10
0.8
0.7
350 kg
dC
50 kg
220000 N/m
2
0
1000
2000
3000
4000
4500
4000
3500
3000
2500
2000
1500
dC 0.6
dW 0.5
20000 N/m
dC [N/(m/s)]
5000
0.4
0.2
0
W
0
1000
2000
3000
4000
5000
dW [N/(m/s)]
dW [N/(m/s)]
Figure 5.10: Quarter car model with sky-hook damper
5.4.3 Technical Realization

By modifying the damper law in Eq. (5.72) to
FD = dW zW dC zC + =
dW zW dC zC
(zW zC ) = dS (zW zC )
zW zC
|
{z
}
dS
(5.77)
the sky-hook damper can be realized by a standard damper in the form of Eq. (5.71). The
new damping parameter dS now nonlinearly depends on the absolute vertical velocities of
the chassis and the wheel dS = dS (zC , zW ). As, a standard damper operates in a dissipative
mode only the damping parameter will be restricted to positive values, dS > 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.
5.5 Nonlinear Force Elements

5.5.1 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.
The equations of motion read as
M zC = FS + FD M g
m zW = FT FS FD m g ,
g = 9.81 m/s2 labels the constant of gravity, M , m are the masses of the chassis
the wheel, FS , FD , FT describe the spring, the damper, and the vertical tire force,
where
and
90
(5.78)
zC
nonlinear spring
nonlinear damper
FS
FD
FS
FD
zW
zR
cT
Figure 5.11: Quarter car model with nonlinear spring and damper characteristics
and the vertical displacements of the chassis
zC
and the wheel
zW
are measured from the
equilibrium position.
In extension to Eq. (5.32) the spring characteristics is modeled by
FS =
FS0
c0 u
c0 u
where
FS0 = M g

1 + kr

1 + kc
u
4ur
2 !
u
4uc
2 !
u<0
(5.79)
u0
is the spring preload, and
u = zW zC
(5.80)
u < 0 marks tension (rebound), and u 0 compression. Two sets of kr , ur and kc , uc dene the spring nonlinearity during rebound and
compression. For kr = 0 and kc = 0 a linear spring characteristics is obtained.
describes the spring travel. Here,
A degressive damper characteristics can be modeled by
FD (v) =
where
d0
d0 v
1 pr v
v <0,
d0 v
1 + pc v
v 0,
denotes the damping constant at
v = 0,
(5.81)
and the damper velocity is dened by
v = zW zC .
(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
pr
91
Vehicle Dynamics
and
pc
make it possible to model the damper nonlinearity dierently in the rebound and
compression mode. A linear damper characteristics is obtained with
pr = 0
and
pc = 0.
The nonlinear spring design in Section 5.3.2 holds for the compression mode. Hence,
c0 = 29 400 N/m, uc = 4u = umax /2 = 0.10/2 = 0.05 and

ur = uc and kr = 0 a simple linear is used in the rebound
using the same data we obtain:
kc = k = 0.1667.
By setting
mode, Fig. 5.12a.
a) Spring
b) Damper
5000
cS = 34300 N/m
6000 c0 = 29400 N/m
ur = 0.05 m
5000 kr = 0
uc = 0.05 m
4000 kc = 0.1667
FD [N/m]
FS [N/m]
7000
2500
d0 = 4200 N/(m/s)
pr = 0.4 1/(m/s)
pc = 1.2 1/(m/s)
3000
2000
-2500
1000
0
-0.1
rebound
u<0
-0.05
compression
u>0
0.05
u [m]
0.1
-5000
-1
rebound
v<0
-0.5
compression
v>0
0.5
v [m/s]
Figure 5.12: Spring and damper characteristics: - - - linear, nonlinear

According to Section 5.3.4 damping coecients optimizing the ride comfort and the ride
safety can be calculated from Eqs. (5.65) and (5.69). For
equivalent linear spring rate,
(dS )C
opt =
(dS )Sopt
2 cS M =
M = 350 kg , m = 50 kg
and
cS = 34 300 N/m which is the

cT = 220 000 N/m we obtain
2 34 300 350 = 4900 N/(m/s) ,

(5.83)
p
(cS + cT ) m = (18 000 + 220 000) 50 = 3570 N/(m/s) .
1
The mean value d0 = 4200N/(m/s) may serve as compromise. With pr = 0.4(m/s)
and
pc = 1.2 (m/s)1 the nonlinearity becomes more intensive in compression than rebound,
Fig. 5.12b.
5.5.2 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.

Compared to the linear model the nonlinear spring and damper characteristics result in
2
2
signicantly reduced peak values for the chassis acceleration (6.0m/s instead of 7.1m/s )
92
Chassis acceleration [m/s2]

10
Wheel load [N]

7000
7.1
6.0
Suspension travel [m]

0.04
6660
6160
6000
0.02
5000
0
4000
-5
3000
-0.02
2000
-10
-15
0.5
1.5
-0.04
linear
nonlinear
1000
0
0.5
time [s]
1.5
-0.06
0.5
time [s]
Figure 5.13: Quarter car model driving with
and for the wheel load (6160
instead of
1.5
time [s]
v = 20 km h
6660 N ).
over a single obstacle
Even the tire lift o at
t 0.25 s
can be avoided. While crossing the bump large damper velocities occur. Here, the degressive damper characteristics provides less damping compared to the linear damper which
increases the suspension travel.
Chassis acceleration [m/s2]

10
5
Wheel load [N]
Suspension travel [m]
7000
0.04
6000
0.02
5000
0
4000
-5
3000
0
-0.02
-0.04
2000
-10
-15
linear,
low damping
1000
0
0.5
time [s]
1.5
-0.06
nonlinear
0.5
time [s]
1.5
-0.08
0.5
1.5
time [s]
Figure 5.14: Results for low damping compared to nonlinear model

A linear damper with a lower damping coecient,
d0 = 3000N/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.
93
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
Fz1
Fx1
mg
Fx2
a2
a1
Fz2
Figure 6.1: Simple vehicle model
If aerodynamic forces (drag, positive and negative lift) are neglected at rst, the equations
of motions in the
where
x-, z -plane
will read as
m v = Fx1 + Fx2 ,
(6.1)
0 = Fz1 + Fz2 m g ,
(6.2)
0 = Fz1 a1 Fz2 a2 + (Fx1 + Fx2 ) h ,
(6.3)
indicates the vehicle's acceleration,
wheel base, and
is the mass of the vehicle,
a1 + a2
is the
is the height of the center of gravity.
These are only three equations for the four unknown forces
Fx1 , Fx2 , Fz1 , Fz2 .
But, if we
insert Eq. (6.1) in Eq. (6.3), we can eliminate two unknowns at a stroke
0 = Fz1 a1 Fz2 a2 + m v h .
94
(6.4)
The equations Eqs. (6.2) and (6.4) can be resolved for the axle loads now
Fz1 = m g
h
a2
m v ,
a1 + a2
a1 + a2
(6.5)
Fz2 = m g
h
a1
+
m v .
a1 + a2
a1 + a2
(6.6)
The static parts
st
Fz1
= mg
a2
,
a1 + a2
st
Fz2
= mg
a1
a1 + a2
(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,
dyn
Fz1
= m g
When accelerating
v > 0,
h
v
,
a1 + a2 g
dyn
Fz2
= +m g
h
v
.
a1 + a2 g
(6.8)
the front axle is relieved as the rear axle is when decelerating
v < 0.
6.1.2 Influence of Grade
v
x
Fx1
mg
Fz1
a1
a2
h
Fx2
Fz2
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 = Fx1 + Fx2 m g sin ,

0 = Fz1 + Fz2 m g cos ,
(6.9)
0 = Fz1 a1 Fz2 a2 + (Fx1 + Fx2 ) h ,
95
Vehicle Dynamics
where
denotes the grade angle. Now, the axle loads are given by
Fz1 = m g cos
h
a2 h tan
m v ,
a1 + a2
a1 + a2
(6.10)
Fz2 = m g cos
a1 + h tan
h
+
m v ,
a1 + a2
a1 + a2
(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 eect 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.
FD1
FD2
FAR
h
mg
Fx1
a2
a1
Fz1
Fx2
Fz2
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 = Fx1 + Fx2 FAR ,

0 = Fz1 FD1 + Fz2 FD2 m g ,
(6.12)
0 = (Fz1 FD1 ) a1 (Fz2 FD2 ) a2 + (Fx1 + Fx2 ) h ,

where
FAR
and
FD1 , FD2
describe the air resistance and the down forces. For the dynamic
axle loads we get
The down forces
Fz1 = FD1 + m g
a2
h
(m v + FAR ) ,
a1 + a2
a1 + a2
(6.13)
Fz2 = FD2 + m g
a1
h
+
(m v + FAR ) .
a1 + a2
a1 + a2
(6.14)
FD1 , FD2
increase the static axle loads, and the air resistance
generates an additional dynamic term.
96
FAR
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
Fz1 0
and
Fz2 0
into account, Eqs. (6.10) and (6.11) will result in
v
a2
cos sin
g
h
and
v
a1
cos sin .
g
h
(6.15)
These two conditions can be combined in one
a1
cos
h
v
+ sin
g
Hence, the maximum achievable accelerations (v
limited by the grade angle
and the
> 0)
position a1 , a2 , h of
a2
cos .
h
(6.16)
and decelerations (v
the center of gravity.
< 0) are
For v
0
the tilting condition Eq. (6.16) results in
a2
a1
tan
h
h
(6.17)
which describes the climbing and downhill capacity of a vehicle.

The presence of aerodynamic forces complicates the tilting condition. Aerodynamic forces
become 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
|Fx1 | Fz1
where the same friction coecient
and
|Fx2 | Fz2
(6.18)
has been assumed at front and rear axle. In the limit
case
Fx1 = Fz1
and
Fx2 = Fz2
(6.19)
the linear momentum in Eq. (6.9) can be written as
m v max = (Fz1 + Fz2 ) m g sin .
(6.20)
Using Eqs. (6.10) and (6.11) one obtains

v
=
g max
That means climbing (v
a friction coecient
> 0, > 0)
tan ||.
cos sin .
or downhill stopping (v
< 0, < 0)
(6.21)
requires at least
97
Vehicle Dynamics
According to the vehicle dimensions and the friction values the maximal acceleration or
deceleration 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
FD1 FD2
+
1 +
mg
mg
FAR
mg
v
g
FD1 FD2
1 +
+
mg
mg
FAR
.
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,
Fx1 = 0
and
Fx2 = Fz2
hold. Then, using
Eq. (6.6) the linear momentum Eq. (6.1) results in

m v R WD = m g
a1
h
v R WD
+
a1 + a2 a1 + a2 g

,
(6.23)
where the subscript R WD indicates the rear wheel drive. Hence, the maximum acceleration
for a rear wheel driven vehicle is given by
v R WD
=
g
By setting
Fx1 = Fz1
and
Fx2 = 0,
h
1
a1 + a2
a1
.
a1 + a2
(6.24)
the maximum acceleration for a front wheel driven
vehicle can be calculated in a similar way. One gets
v F WD
=
g
a2
,
h
a1 + a2
1+
a1 + a2
where the subscript F WD denotes front wheel drive. Depending on the parameter
a2 and h the accelerations may be limited by the tilting condition vg ah2 .
(6.25)
, a1 ,
The maximum accelerations of a single axle driven vehicle are plotted in Fig. 6.4. For rear
a2 /(a1+a2 ) which describes the static axle load

distribution is in the range of 0.4 a2 /(a1+a2 ) 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 a2 /(a1 +a2 ) 0.60 which produces accelerations in
the range of 0.45 v/g
0.49. Hence, rear wheel driven vehicles can accelerate much
wheel driven passenger cars, the parameter
faster than front wheel driven vehicles.
98
range of load distribution
v/g
0.8
FWD
RWD
FWD
0.6
0.4
RWD
0.2
0
0.2
0.4
0.6
Figure 6.4: Single axle driven passenger car:
0.8
a2 / (a1+a2)
= 1, h = 0.55 m, a1 +a2 = 2.5 m
6.3.2 Braking at Single Axle

If only the front axle is braked, in the limit case
Fx1 = Fz1
and
Fx2 = 0
will hold. With
Eq. (6.5) one gets from Eq. (6.1)

m v F WB = m g
h
v F WB
a2
a1 + a2 a1 + a2 g

,
(6.26)
where the subscript F WB indicates front wheel braking. Then, the maximum deceleration
is given by
v F WB
=
g
If only the rear axle is braked (Fx1
deceleration
h
1
a1 + a2
a2
.
a1 + a2
= 0, Fx2 = Fz2 ),
v R WB
=
g
h
1+
a1 + a2
(6.27)
one will obtain the maximum
a1
,
a1 + a2
(6.28)
where the subscript R WB denotes a braked rear axle. Depending on the parameters
v
and h, the decelerations may be limited by the tilting condition
ah1 .
g
, a1 ,
a2 ,
The maximum decelerations of a single axle braked vehicle are plotted in Fig. 6.5. For
passenger cars the load distribution parameter
to
0.6.
a2 /(a1 +a2 ) usually covers the range of 0.4

v/g
= 0.51 to v/g
= 0.77
If only the front axle is braked, decelerations from
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.
99
Vehicle Dynamics
range of
load
distribution
v/g
-0.2
FWB
-0.4
-0.6
-0.8
-1
RWB
0
0.2
0.4
0.6
Figure 6.5: Single axle braked passenger car:
0.8
a2 / (a1+a2)
= 1, h = 0.55 m, a1 +a2 = 2.5 m
6.3.3 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
Fx1
Fx2
v
=
+
.
g
mg
mg
(6.29)
A certain acceleration or deceleration can only be achieved by dierent combinations of

the longitudinal forces
Fx1
and
Fx2 .
According to Eq. (6.19) the longitudinal forces are
limited by wheel load and friction.

The optimal combination of
Fx1
and
Fx2
will be achieved, when front and rear axle have
the same skid resistance.
Fx1 = Fz1
and
Fx2 = Fz2 .
(6.30)
With Eq. (6.5) and Eq. (6.6) one obtains
and
Fx1
=
mg
Fx2
=
mg
a2
v
h
g
a1
v
+
h
g
h
a1 + a2
(6.31)
h
.
a1 + a2
(6.32)
With Eq. (6.31) and Eq. (6.32) one gets from Eq. (6.29)
v
= ,
g
where it has been assumed that
Fx1
and
Fx2
(6.33)
have the same sign. Finally, if Eq. (6.33 is
inserted in Eqs. (6.31) and (6.32) one will obtain
Fx1
v
=
mg
g
100
a2
v
h
g
h
a1 + a2
(6.34)
and
Fx2
v
=
mg
g
Depending on the desired acceleration
a1
v
+
h
g
h
.
a1 + a2
(6.35)
v > 0 or deceleration v < 0, the longitudinal forces
B2/mg
that grant the same skid resistance at both axles can be calculated now.
braking
Fx1/mg
-a1/h
dFx2
dFx1 0
-1
-2
B1/mg
a =1.15
1
driving
a =1.35
2
h=0.55
=1.20
tilting limits
Fx2/mg
a2/h
Figure 6.6: Optimal distribution of driving and braking forces

Fig. 6.6 shows the curve of optimal drive and brake forces for typical passenger car values.
At the tilting limits
v/g
= a1 /h and v/g
= +a2 /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
Fx1

a2
v
h
mg
=
2
v
h
g a1 + a2
d
g
(6.36)
Fx2

a1
v
h
mg
=
+2
.
v
h
g a1 + a2
d
g
(6.37)
and
101
Vehicle Dynamics
For
v/g
=0
the initial gradient remains as

d Fx2
a1
.
=

d Fx1 0
a2
(6.38)
6.3.4 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. 6.7.
limitation
Fx2/mg
Fx2/mg
constant
distribution
Fx1/mg
Fx2/mg
Fx1/mg
Fx1/mg
reduction
Figure 6.7: Dierent 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 center 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.
6.3.5 Anti-Lock-Systems
On hard braking maneuvers large longitudinal slip values occur. Then, the stability and/or
steerability 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 dierentiation.
The rolling condition is fullled with a longitudinal slip of
rD = x
102
sL = 0.
Then
(6.39)
holds, where
rD
labels the dynamic tire radius and
names the longitudinal acceleration
of the vehicle. According to Eq. (6.21), the maximum acceleration/deceleration of a vehicle

depends on the friction coecient,
|
x| = g .
For a given friction coecient
a simple
control law can be realized for each wheel

||
1
|
x| .
rD
(6.40)
Because no reliable possibility to determine the local friction coecient 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
held,
or
decreased
increased,
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. 6.8 consists of ve rigid bodies. The body has three degrees
of freedom: Longitudinal motion
z1
and
z2
xA ,
vertical motion
zA
and pitch
A .
The coordinates
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:
x1 = x1 (z1 ) , 1 = 1 (z1 ) ;
x2 = x2 (z2 ) , 2 = 2 (z2 ) .
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
drive torques
(6.41)
FF 1
and
FF 2 .
At the wheels
MA1 , MA2 and brake torques MB1 , MB2 , longitudinal forces Fx1 , Fx2 and
Fz1 , Fz2 apply. The brake torques are directly supported by the wheel
the wheel loads
bodies, whereas the drive torques are transmitted by the drive shafts to the vehicle body.
103
Vehicle Dynamics
zA
xA
FF1
MA1
z1
MB1
MA1
hR
R1
MB1
Fz1
z2
MB2
MA2
Fx1
R
a1
FF2
MA2
R2
a2
MB2
Fz2
Fx2
Figure 6.8: Simple vehicle model
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
v0A,0
x A
0
= 0 + 0 ;
0
zA
0
= A .
0
0A,0
(6.42)
At small rotational motions of the body one gets for the velocities of the wheel bodies
and wheels
x1
z1
v0RK1 ,0 = v0R1 ,0
x A
0
hR A
+
0
= 0 + 0 +
0
zA
a1 A
x2
z2
v0RK2 ,0 = v0R2 ,0
x A
0
hR A
+
0
= 0 + 0 +
0
zA
+a2 A
z1
0
z1
z2
0
z2
(6.43)
(6.44)
The angular velocities of the wheel bodies and wheels are obtained from
0RK1 ,0
0
0
= A + 1
0
0
and
0R1 ,0
0
0
0
= A + 1 + R1
0
0
0
(6.45)
104
as well as
0
0
= A + 2
0
0
0RK2 ,0
0R2 ,0
and
0
0
0
= A + 2 + R2
0
0
0
(6.46)
Introducing a vector of generalized velocities
z =
x A zA A 1 R1 2 R2
T
(6.47)
the velocities and angular velocities given by Eqs. (6.42), (6.43), (6.44), (6.45), and (6.46)
can be written as
v0i =
7
X
v0i
j=1
zj
zj
and
0i =
7
X
0i
j=1
zj
zj
(6.48)
6.4.2 Equations of Motion

The partial velocities
0i
v0i
and partial angular velocities
for the ve bodies
zj
zj
and for the seven generalized speeds
j = 1(1)7
partial velocities
bodies
chassis
mA
wheel body
front
mRK1
wheel
front
mR1
wheel body
rear
mRK2
wheel
rear
mR2
x A
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
zA
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
A
0
0
0
hR
0
a1
hR
0
a1
hR
0
a2
hR
0
a2
z1
0
0
0
x1
z1
0
1
x1
z1
0
1
0
0
0
0
0
0
i = 1(1)5
are arranged in the tables 6.1 and 6.2.
v0i /zj
R1
z2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
x2
0
z2
0
0
0
1
x2
0
z2
0
0
0
1
applied forces
R2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Fie
0
0
FF 1 +FF 2 mA g
0
0
FF 1 mRK1 g
Fx1
0
Fz1 mR1 g
0
0
FF 2 mRK2 g
Fx2
0
Fz2 mR2 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 matrix
and the components of the vector of generalized forces and torques
Q can
be calculated.
M (i, j) =
T
5
X
v0k
k=1
zi
T
5
X
v0k
0k
0k
mk
+
k
;
zj
zj
zi
k=1
i, j = 1(1)7 ;
(6.49)
105
Vehicle Dynamics
partial angular velocities

bodies
chassis
wheel body
front
RK1
wheel
front
R1
wheel body
rear
RK2
wheel
rear
R2
x A
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
zA
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
z1
0
0
0
0
R1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
z1
0
0
1
z1
0
0
0
0
0
0
0
0i /zj
applied torques
z2 R2
Mie
0
0
0
0
0 MA1MA2 a1 FF 1 +a2 FF 2
0
0
0
0
0
0
0
0
MB1
0
0
0
0
0
0
0
0
MA1 MB1 R Fx1
0
0
0
0
0
0
2
0
M
B2
z2
0
0
0
0
0
0
2
1
M
M
A2
B2 R Fx2
z2
0
0
0
Table 6.2: Partial angular velocities and applied torques
Q(i) =
T
5
X
v0k
k=1
zi
Fke
T
5
X
0k
k=1
zi
Mke ;
i = 1(1)7 .
(6.50)
Then, the equations of motion for the plane vehicle model are given by
M z = Q .
(6.51)
6.4.3 Equilibrium
With the abbreviations
m1 = mRK1 + mR1 ;
m2 = mRK2 + mR2 ;
mG = mA + m1 + m2
(6.52)
and
h = hR + R
(6.53)
The components of the vector of generalized forces and torques read as
Q(1) = Fx1 + Fx2 ;

Q(2) = Fz1 + Fz2 mG g ;
(6.54)
Q(3) = a1 Fz1 + a2 Fz2 h(Fx1 + Fx2 ) + a1 m1 g a2 m2 g ;

Q(4) = Fz1 FF 1 +
x1
z1
Fx1 m1 g +
Q(5) = MA1 MB1 R Fx1 ;
106
1
(MA1
z1
R Fx1 ) ;
(6.55)
Q(6) = Fz2 FF 2 +
x2
z2
Fx2 m2 g +
2
(MA2
z2
R Fx2 ) ;
(6.56)
Q(7) = MA2 MB2 R Fx2 .

Without drive and brake forces
MA1 = 0 ;
MA2 = 0 ;
MB1 = 0 ;
MB2 = 0
(6.57)
from Eqs. (6.54), (6.55) and (6.56) one gets the steady state longitudinal forces, the spring
preloads, and the wheel loads
0
Fx1
= 0;
FF0 1 =
0
Fz1
0
Fx2
= 0;
b
a+b
= m1 g +
a
a+b
FF0 2 =
mA g ;
a2
a1 +a2
0
Fz2
mA g ;
mA g ;
= m2 g +
(6.58)
a1
a1 +a2
mA g .
6.4.4 Driving and Braking

Assuming that on accelerating or decelerating the vehicle the wheels neither slip nor lock,
R R1 = x A hR A +
R R2 = x A hR A +
x1
z1
x2
z2
z1 ;
(6.59)
z2
hold. In steady state the pitch motion of the body and the vertical motion of the wheels
reach constant values
A = Ast = const. ,
z1 = z1st = const. ,
z2 = z2st = 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
a
a
mG xA = Fx1
+ Fx2
;
a
a
0 = Fz1
+ Fz2
;
hR (m1 +m2 ) xA + R1
x
A
R
+ R2
x
A
R
a
a
a
a
= a Fz1
+ b Fz2
(hR + R)(Fx1
+ Fx2
);
(6.62)
x1
z1
x2
z2
m1 xA +
m2 xA +
1
z1
2
z2
R1 xRA
a
Fz1
FFa 1
x1
z1
a
Fx1
R1
x
A
R
a
= MA1 MB1 R Fx1
;
R2
x
A
R
a
= Fz2
FFa 2 +
R2
x
A
R
a
= MA2 MB2 R Fx2
;
x2
z2
a
Fx2
+
1
(MA1
z1
2
(MA2
z2
a
R Fx2
);
a
R Fx1
)
(6.63)
(6.64)
107
Vehicle Dynamics
where the steady state spring forces, longitudinal forces, and wheel loads have been separated into initial and acceleration-dependent terms
a
0
st
;
+ Fxi
= Fxi
Fxi
Fzist = Fzi0 + Fzia ;
FFsti = FF0 i + FFa i ;
i = 1, 2 .
(6.65)
a
With given torques of drive and brake the vehicle acceleration x
A , the wheel forces Fx1
,
a
a
a
a
a
Fx2 , Fz1 , Fz2 and the spring forces FF 1 , FF 2 can be calculated from Eq. (6.62), Eq. (6.63)
and Eq. (6.64)
Via the spring characteristics which have been assumed as linear the accelerationdependent forces also cause a vertical displacement and pitch motion of the body besides
the vertical motions of the wheels,
FFa 1
FFa 2
a
Fz1
a
Fz2
Especially the pitch of the vehicle
= cA1 z1a ,
= cA2 z2a ,
= cR1 (zAa a Aa + z1a ) ,
= cR2 (zAa + b Aa + z2a ) .
(6.66)
Aa 6= 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. 6.9.
pitch pole
x-, z- motion of the contact points

during compression and rebound
Figure 6.9: 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.
108
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 vanishing 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
vy = eTy v0P = 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
a
a
and
tan 2 =
,
(7.2)
R
R+s
where s labels the track width and a denotes the wheel base. Eliminating the curve radius
R, we get
a
a tan 1
tan 2 =
or
tan 2 =
.
(7.3)
a
a + s tan 1
+s
tan 1
a
A
a
The deviations 42 = 2 2 of the actual steering angle 2 from the Ackermann steering
A
angle 2 , which follows from Eq. (7.3), are used, especially on commercial vehicles, to
tan 1 =
judge the quality of a steering system.

At a rotation around the momentary pole
every point of the vehicle. The angle
M,
the direction of the velocity is xed for
between the velocity vector
and the longitudinal
axis of the vehicle is called side slip angle. The side slip angle at point
tan P =
where
denes the distance of
x
R
P to
or
tan P =
x
tan 1 ,
a
is given by
(7.4)
the inner rear wheel.
109
Vehicle Dynamics
v
P
P
a
x
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
Rmax ,
whereas a point on the inner side of the vehicle at the location of the rear
axle will run on the minimum radius
Rmin .
Hence, it holds
2
Rmax
= (Rmin + b)2 + (a + f )2 ,
where
a, b are the wheel base and the width of the vehicle, and f
(7.5)
species the distance from
the front of the vehicle to the front axle. Then, the space requirement
can be specied as a function of the cornering radius
4R = Rmax Rmin =
The space requirement
4R
Rmin
110
2.5
for a given vehicle dimension
(Rmin + b)2 + (a + f )2 Rmin .
(7.6)
of a typical passenger car and a bus is plotted in Fig. 7.3
Rmin = 5.0 m, a bus requires

needs only 1.5 times the width.
versus the minimum cornering radius. In narrow curves

space of
4R = Rmax Rmin
times the width, whereas a passenger car
ma
Rmin
Figure 7.2: Space requirement
bus: a=6.25 m, b=2.50 m, f=2.25 m

car: a=2.50 m, b=1.60 m, f=1.00 m
R [m]
5
4
3
2
1
0
10
20
30
R min [m]
40
50
Figure 7.3: Space requirement of a typical passenger car and bus
111
Vehicle Dynamics
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
x0 , y0 , z0
= y
R
of the vehicle relative to the track xed frame

to the rear axle center
r02,0
and the rotation matrix
(7.7)
A02
Here, the tire radius
is dened by the position vector
cos sin 0
cos 0 .
= sin
0
0
1
is considered to be constant, and
(7.8)
x, y
as well as the yaw angle
are generalized coordinates.
a
b
A1
y2
x2
A2
y0
y3
A3
x0
Figure 7.4: Kinematic model with trailer

The position vector
r01,0 = r02,0 + A02 r21,2
112
with
r21,2
a
= 0
0
(7.9)
A01 = A02 A21
with
cos sin 0
cos 0
= sin
0
0
1
A21
describe the position and the orientation of the front axle, where
wheel base and
a = const
(7.10)
labels the
the steering angle.
The position vector
r03,0 = r02,0 + A02 r2K,2 + A23 rK3,3

with
r2K,2
b
= 0
0
(7.11)
c
= 0
0
(7.12)
cos sin 0
cos 0
= sin
0
0
1
(7.13)
rK3,2
and
A03 = A02 A23
with
A23
dene the position and the orientation of the trailer axis, with
between vehicle and trailer, and
coupling point
b, c
labeling the bend angle
marking the distances from the rear axle
and from the coupling point
to the trailer axis
to the
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
v02,2
vx2
= 0 .
0
(7.14)
The time derivative of Eq. (7.7) results in
v02,0 = r02,0
x
= y .
0
The transformation of Eq. (7.14) into the system
v02,0 = A02 v02,2 = A02
(7.15)
vx2
cos vx2
0 = sin vx2
0
0
(7.16)
113
Vehicle Dynamics
compared to Eq. (7.15) results in two rst order dierential equations for the position
coordinates
and
y
x = vx2 cos ,
y = vx2 sin .
(7.17)
The velocity at the front axle follows from Eq. (7.9)
v01,0 = r01,0 = r02,0 + 02,0 A02 r21,2 .

x2 , y2 , z2 we obtain
vx2
0
a
vx2
= 0 + 0 0 = a .
0
0
0
| {z }
| {z } | {z }
v02,2
02,2
r21,2
(7.18)
Transformed into the vehicle xed system
v01,2
(7.19)
The unit vectors
ex1,2
cos
= sin
0
and
ey1,2
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,
eTy1,2 v01,2 = sin vx2 + cos a = 0 .
(7.21)
Whereas in longitudinal direction the velocity
eTx1,2 v01,2 = cos vx2 + sin a = vx1
(7.22)
remains. From Eq. (7.21) a rst order dierential equation follows for the yaw angle
vx2
tan .
a
(7.23)
7.1.4.3 Entering a Curve

In analogy to Eq. (7.2) the steering angle
or with
k = 1/R
can be related to the current track radius
to the current track curvature
tan =
a
= ak .
R
(7.24)
Then, the dierential equation for the yaw angle reads as
= vx2 k .
With the curvature gradient
k = k(t) = kC
114
(7.25)
t
,
T
(7.26)
the entering of a curve is described as a continuous transition from a straight line with
k=0
the curvature
into a circle with the curvature
k = kC .
The yaw angle of the vehicle can be calculated by simple integration now
vx2 kC t2
,
T
2
angle, (t = 0) = 0,
(t) =
where at time
t=0
a vanishing yaw
(7.27)
has been assumed. Then, the
position of the vehicle follows with Eq. (7.27) from the dierential equations Eq. (7.17)
Zt=T
x = vx2

cos
vx2 kC t2
T
2
Zt=T

dt ,
y = vx2
t=0

sin
vx2 kC t2
T
2

dt .
(7.28)
t=0
At constant vehicle speed,
vx2 = 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
angles are necessary for driving through the curve, the
tan-function can be approximated
by its argument, and
= (t) a k = a kC
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 dierentiation of the position vector
Eq. (7.11)
v03,0 = r03,0 = r02,0 + 02,0 A02 r23,2 + A02 r23,2 .

The velocity
r02,0 = v02,0
and the angular velocity
02,0
(7.30)
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
r23,2 = r2K,2 + A23 rK3,3

where
r2K,2
and
rK3,3
b c cos
c sin ,
=
0
(7.31)
are dened in Eq. (7.12). The time derivative of Eq. (7.31) results
in

c cos
c sin
0
= 0 c sin = c cos .
0
0
{z
}
| {z } |
23,2
A23 rK3,3
r23,2
(7.32)
x2 , y2 , z2 now

vx2
0
b c cos
c sin
vx2 + c sin (+
)
c sin + c cos = b c cos (+

)
.
= 0 + 0
0
0
0
0
{z
} |
{z
}
| {z } | {z } |
v02,2
02,2
r23,2
r23,2
Eq. (7.30) is transformed into the vehicle xed frame
v03,2
(7.33)
115
Vehicle Dynamics
The longitudinal and lateral direction at the trailer axle are dened by the unit vectors
ex3,2
cos
= sin
0
ey3,2
and
sin
= cos .
0
(7.34)
At the trailer axis the lateral velocity must also vanish

)
+ cos b c cos (+
)
eTy3,2 v03,2 = sin vx2 + c sin (+

= 0,
(7.35)
whereas in longitudinal direction the velocity

)
eTx3,2 v03,2 = cos vx2 + c sin (+

+ sin b c cos (+
= vx3
(7.36)
remains. If Eq. (7.23) is inserted into Eq. (7.35) now, one will get a rst order dierential
equation for the bend angle
vx2
=
a
a
sin +
c
b
cos + 1 tan
c

.
(7.37)
The dierential equations Eq. (7.17) and Eq. (7.23) describe position and orientation
x0 , y0
within the
plane. The position of the trailer relative to the vehicle follows from
Eq. (7.37).
7.1.4.5 Course Calculations
front axle
rear axle
trailer axle
20
[m]
10
0
-30
-20
-10
10
20
30 [m] 40
50
60
30
[o]
front axle steering angle
20
10
0
10
15
20
[s]
25
30
Figure 7.5: Entering a curve

For a given set of vehicle parameters
velocity,
vx2 = vx2 (t)
a, b , c ,
and predened time functions of the vehicle
and the steering angle,
= (t),
the course of vehicle and trailer
can be calculated by numerical integration of the dierential 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.
116
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
v0C,B
where
v cos
= v sin
0
and
0F,B
0
= 0 ,
denotes the side slip angle of the vehicle measured at the center of gravity. The
angular velocity of a vehicle cornering with constant velocity

is given by
=
where
(7.38)
on an at horizontal road
v
,
R
(7.39)
denotes the radius of curvature.
Fx2
a2
Fy2
a1
xB
Fx1
R
yB
Fy1
Figure 7.6: Cornering resistance
In the body xed reference frame, linear and angular momentum result in

v2
m sin
= Fx1 cos Fy1 sin + Fx2 ,
R
2

v
m
cos
= Fx1 sin + Fy1 cos + Fy2 ,
R
0 = a1 (Fx1 sin + Fy1 cos ) a2 Fy2 ,

(7.40)
(7.41)
(7.42)
117
Vehicle Dynamics
where
denotes the mass of the vehicle,
Fx1 , Fx2 , Fy1 , Fy2
are the resulting forces in
longitudinal 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 dierential to the front and rear axle. Then,
in steady state condition we obtain
Fx1 = k FD
where
FD
is the driving force and by
k=0
and
Fx2 = (1 k) FD ,
dierent driving conditions
rear wheel drive
0<k<1
all wheel drive
k=1
front wheel drive
(7.43)
can be modeled:
Fx1 = 0, Fx2 = FD
Fx1
k
=
Fx2
1k
Fx1 = FD , Fx2 = 0
If we insert Eq. (7.43) into Eq. (7.40) we will get

k cos + (1k) FD
k sin FD +
sin Fy1
cos Fy1 +
Fy2
mv 2
=
sin ,
R
mv 2
=
cos ,
R
(7.44)
a1 k sin FD + a1 cos Fy1 a2 Fy2 = 0 .

These equations can be resolved for the driving force
FD
a2
cos sin sin cos
mv 2
a1 + a2
=
.
k + (1 k) cos
R
(7.45)
The driving force will vanish, if
a2
cos sin = sin cos
a1 + a2
or
a2
tan = tan
a1 + a2
(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
a2
angle. Then, due to tan <
tan a driving force FD > 0 is needed to overcome the
a1 +a2
'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.
118
2 1
m ay
h2
mg
h1
F yL
F yR
FzL
s/2
s/2
FzR
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
s
= m ay (h1 + h2 ) + m g [(h1 + h2 )1 + h2 2 ] ,
2
where ay describes the lateral acceleration, m is the sprung mass, and small
the axle and the body were assumed, 1 1, 2 1.
(FzL FzR )
(7.47)
roll angles of
On a left-hand tilt, the right tire raises
T
FzR
= 0,
(7.48)
whereas the left tire carries the complete vehicle weight
T
FzL
= mg .
(7.49)
Using Eqs. (7.48) and (7.49) one gets from Eq. (7.47)
s
aTy
h2
2
=
1T
2T .
g
h1 + h2
h1 + h2
(7.50)
T
rises above the limit ay . Roll
T
T
of axle and body reduce the overturning limit. The angles 1 and 2 can be calculated
from the tire stiness cR and the roll stiness of the axle suspension.
The vehicle will turn over, when the lateral acceleration
ay
119
Vehicle Dynamics
If the vehicle drives straight ahead, the weight of the vehicle will be equally distributed
to both sides
stat
stat
FzR
= FzL
=
1
mg .
2
(7.51)
With
T
stat
FzL
= FzL
+ 4Fz
(7.52)
and Eqs. (7.49), (7.51), one obtains for the increase of the wheel load at the overturning
limit
4Fz =
1
mg .
2
(7.53)
Then, the resulting tire deection follows from
4Fz = cR 4r ,
where
cR
(7.54)
is the radial tire stiness.
Because the right tire simultaneously rebounds with the same amount, for the roll angle
of the axle
2 4r = s 1T
or
1T =
mg
2 4r
=
s
s cR
(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
cW 2 = m ay h2 + m g h2 (1 + 2 ) ,
(7.56)
cW names the roll stiness of the body suspension. In particular, at the overturning
ay = aTy
aTy
mgh2
mgh2
T
2 =
+
1T
(7.57)
g cW mgh2
cW mgh2
where
limit
T
applies. Not allowing the vehicle to overturn already at ay = 0 demands a minimum of
min
roll stiness cW > cW
= mgh2 . With Eqs. (7.55) and (7.57) the overturning condition
Eq. (7.50) reads as
(h1 + h2 )
aTy
aTy
s
1
1
1
1
h2
,
=
(h1 + h2 ) h2
g
2
cR
g cW 1
cW 1 cR
(7.58)
where, for abbreviation purposes, the dimensionless stinesses
cR
cR = m g
s
and
cW =
cW
m g h2
(7.59)
have been used. Resolved for the normalized lateral acceleration
aTy
=
g
120
s
2
h2
h1 + h2 +
cW 1
1
cR
(7.60)
roll angle =T1+2

T
overturning limit ay
0.6
20
0.5
15
0.4
0.3
10
0.2
5
0.1
0
0
10
20
normalized roll stiffness c W
*
0
10
20
normalized roll stiffness c W
*
Figure 7.8: Tilting limit for a typical truck at steady state cornering
remains.
m = 13 000 kg . The radial stiness

width can be set to s = 2 m. The values
At heavy trucks, a twin tire axle may be loaded with
cR = 800 000 N/m, and the track

h1 = 0.8 m and h2 = 1.0 m hold at maximal load. These values produce the results shown
in Fig. 7.8. Even with a rigid body suspension cW , the vehicle turns over at a lateral
acceleration of ay 0.5 g . Then, the roll angle of the vehicle solely results from the tire
of one tire is
deection.
At a normalized roll stiness of
cW = 5,
the overturning limit lies at
ay 0.45 g
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
ay ,
centrifugal forces
will be applied to the single masses. At the simple roll model in Fig. 7.9, these are the
forces
m A ay
and
m R ay ,
where
Through the centrifugal force
mA
m A ay
names the body mass and
the wheel mass.
applied to the body at the center of gravity, a torque
is generated, which rolls the body with the angle

of the tires
mR
and leads to an opposite deection
z1 = z2 .
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
Fy1 , Fz1 , Fy2 , Fz2 ,
FF 1 , FF 2
and tire forces
are approximated by linear spring elements with the constants
cA
and
121
Vehicle Dynamics
b/2
b/2
zA
mA a y
A
SA
yA
FF1
FF2
h0
z2
z1
mR a y
mR a y
S2
r0
y2
Q2
S1
Q1
F y2
Fy2
F z1
y1
F y1
Figure 7.9: Simple vehicle roll model
cQ , cR .
The work
potential
V.
of these forces can be calculated directly or using
W = V
via the
At small deections with linearized kinematics one gets
W = mA ay yA
mR ay (yA + hR A + y1 )2 mR ay (yA + hR A + y2 )2
21 cA z12
1
2
cA z22
(7.62)
21 cS (z1 z2 )2
21 cQ (yA + h0 A + y1 + r0 1 )2 12 cQ (yA + h0 A + y2 + r0 2 )2
2
2
21 cR zA + 2b A + z1 21 cR zA 2b A + z2 ,
where the abbreviation
hR = h0 r0
has been used, and
cS
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
y1 =
the work
y
z1 ,
z
1 =
1
z
and
y2 =
y
z2 ,
z
2 =
2
z
(7.63)
can be described as a function of the position vector
y = [ yA , zA , A , z1 , z2 ]T .
(7.64)
W = W (y)
(7.65)
Due to
the principle of virtual work Eq. (7.61) leads to
W =
122
W
y = 0 .
y
(7.66)
Because of
y 6= 0,
a system of linear equations in the form of
Ky = b
results from Eq. (7.66). The matrix
2 cQ
K = 2 cQ h0
yQ c
z Q
Q
yz cQ
Q
yz cQ
cR
Q
b
c +h0 yz cQ
2 R
cA + cS + cR
2b cR h0 yz cQ
and
are given by
y Q
c
z Q
Q
b
c +h0 yz cQ
2 R
cR
and the vector
2 cQ h0
2 cR
cR
(7.67)
b =
cS
mA + 2 mR
0
(m1 + m2 ) hR
mR y/z
mR y/z
cR
Q
y
b
2 cR h0 z cQ
cS
cA + cS + cR
(7.68)
ay .
(7.69)
The following abbreviations have been used:
y
y Q
=
+ r0
,
z
z
z
cA

= cA + cQ
y
z
2
,
c =
2 cQ h20
2
b
+ 2 cR
.
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)
b)
roll center
1 0
roll center
2 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
123
Vehicle Dynamics
sportive vehicles axle kinematics are employed, where the wheels are rotated around the
1 = 1 (z1 ) and 2 = 2 (z2 ). Hereby, a camber

1 0 and 2 0. Fig. 7.10 b. By the rotation of
longitudinal axis at jounce and rebound,

compensation can be achieved with
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 axis
roll center front
roll center rear
Figure 7.11: Roll axis

The 'roll center' can be constructed from the lateral motion of the wheel contact points
Q1
and
Q2 ,
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

-TT
+TT
PR0+P
PF0+P
PF0-P
PR0+PR
PR0-P
PF0+PF
PR0-PR
PF0-PF
Figure 7.12: Wheel loads for a exible and a rigid chassis

The roll angle of a vehicle during cornering depends on the roll stiness of the axle and
on the position of the roll center. Dierent axle layouts at the front and rear axle may
result in dierent roll angles of the front and rear part of the chassis, Fig. 7.12.
124
On most passenger cars the chassis is rather sti. Hence, front and rear part of the chassis
are forced by an internal torque to an overall chassis roll angle. This torque aects the
wheel loads and generates dierent wheel load dierences 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 aected.
7.3 Simple Handling Model

7.3.1 Modeling Concept
x0
a2
a1
y0
Fy2
x2
xB
Fy1
y1
y2
yB
x1
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
xB , yB , zB
vC,B
where
v cos
= v sin ,
0
denotes the side slip angle, and
(7.71)
is the magnitude of the velocity.
125
Vehicle Dynamics
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
ex1 ,B
cos
= sin ,
0
and
ey1 ,B
sin
= cos ,
0
ex2 ,B
where
and
a1
and
a2
1
= 0 ,
0
v01,B
ey2 ,B
0
= 1 ,
0
v cos
= v sin + a1
0
(7.72)
v02,B
v cos
= v sin a2 ,
0
(7.73)
are the distances from the center of gravity to the front and rear axle,
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 permitted. At small lateral slips, the lateral force can be approximated by a linear
approach
Fy = cS sy ,
where
cS
is a constant depending on the wheel load
(7.74)
Fz , and the lateral slip sy
is dened by
Eq. (3.88). Because the vehicle is neither accelerated nor decelerated, the rolling condition
is fullled at each wheel
rD = eTx v0P .
Here,
rD
is the dynamic tire radius,
v0P
the contact point velocity, and
(7.75)
ex
the unit vector
in longitudinal direction. With the lateral tire velocity
vy = eTy v0P
(7.76)
and the rolling condition Eq. (7.75), the lateral slip can be calculated from
eTy v0P
,
sy = T
| ex v0P |
with
ey
(7.77)
labeling the unit vector in the lateral direction direction of the tire. So, the lateral
forces are given by
Fy1 = cS1 sy1 ; Fy2 = cS2 sy2 .
(7.78)
7.3.4 Lateral Slips

With Eq. (7.73), the lateral slip at the front axle follows from Eq. (7.77):
sy1 =
126
+ sin (v cos ) cos (v sin + a1 )
.
| cos (v cos ) + sin (v sin + a1 )
|
(7.79)
The lateral slip at the rear axle is given by
sy2 =
The yaw velocity of the vehicle
v sin a2
.
| v cos |
the side slip angle
(7.80)
and the steering angle
are
considered to be small
| a1 | |v| ;
| | 1
| a2 | |v|
(7.81)
|| 1 .
(7.82)
and
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
sy1 =
a1
v
+
|v|
|v|
(7.84)
a2
,
|v|
(7.85)
and
sy2 = +
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
1, Eq. (7.71)
v
vC,B = |v| .
0
of motion, For small side slip angles
can be approximated by
(7.86)
The angular velocity is given by
0F,B
0
= 0 .
If the vehicle accelerations are also expressed in the vehicle xed frame
(7.87)
xF , yF , zF ,
one
v = const and with neglecting small higher-order terms
0
= 0F,B vC,B + v C,B = v + |v| .
(7.88)
0
will nd at constant vehicle speed
aC,B
127
Vehicle Dynamics
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
= Fy1 + Fy2 ,
m (v + |v| )
where, due to the small steering angle, the term
and
Fy1 cos
(7.91)
has been approximated by
describes the vehicle mass. With Eq. (7.90) the angular momentum yields
= a1 Fy1 a2 Fy2 ,
where
Fy1 ,
(7.92)
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 dierential equations
cS1
m |v|
a1 cS1

a2
cS2
v
|v|
m |v|
|v|

a1
v
a2
a2 cS2

+
,
|v|
|v|
|v|
a1
v
+
|v|
|v|
(7.93)
(7.94)
which can be written in the form of a state equation
cS1 + cS2
m |v|
a2 cS2 a1 cS1
| {z }
x
|

v
a2 cS2 a1 cS1
v cS1

m |v||v|
|v|
|v| m |v|
+
| {z } v a1 cS1
a21 cS1 + a22 cS2
x
|v|
|v|
{z
}
|
{z
A
B

.
|{z}
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 methods of system dynamics.
7.3.6 Stability
7.3.6.1 Eigenvalues
The homogeneous state equation
x = A x
128
(7.96)
describes the eigen-dynamics. If the approach
xh (t) = x0 et
(7.97)
is inserted into Eq. (7.96), the homogeneous equation will remain
( E A) x0 = 0 .
One gets non-trivial solutions
x0 6= 0
(7.98)
for
det | E A| = 0 .
The eigenvalues
(7.99)
provide information concerning the stability of the system.
7.3.6.2 Low Speed Approximation

The state matrix
cS1 + cS2
m |v|
=
Av0
v
a2 cS2 a1 cS1
m |v||v|
|v|
2
2
a1 cS1 + a2 cS2
|v|
(7.100)
v 0.
The matrix in
a21 cS1 + a22 cS2

.
|v|
(7.101)
approximates the eigen-dynamics of vehicles at low speeds,

Eq. (7.100) has the eigenvalues
1v0 =
cS1 + cS2
m |v|
and
2v0 =
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
Av
=
a2 cS2 a1 cS1
|v|
.
(7.102)
Using Eq. (7.102) one receives from Eq. (7.99) the relation
2v +
v a2 cS2 a1 cS1
= 0
|v|
(7.103)
129
Vehicle Dynamics
with the solutions
1,2v
When driving forward with
v > 0,
r
v a2 cS2 a1 cS1
.
=
|v|
(7.104)
the root argument will be positive, if
a2 cS2 a1 cS1 < 0
(7.105)
holds. Then however, one eigenvalue is positive, and the system is unstable. Two zeroeigenvalues
1 = 0
and
2 = 0
are obtained for
a1 cS1 = a2 cS2 .
(7.106)
The driving behavior is indierent then. Slight parameter variations, however, can lead
to an unstable behavior. With
a2 cS2 a1 cS1 > 0

and
v > 0
or
a1 cS1 < a2 cS2
(7.107)
the root argument in Eq. (7.104) becomes negative. Then, the eigenvalues
are imaginary, and disturbances lead to undamped vibrations. To avoid instability, highspeed 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!
7.3.7 Steady State Solution

7.3.7.1 Side Slip Angle and Yaw Velocity
= 0 , a stable system reaches steady state after a certain time.

x st = 0, the state equation Eq. (7.95) is reduced to a system of linear
At a given steering angle

With
xst = const.
or
equations
A xst = B u .
With the elements from the state matrix
and the vector
two equations to determine the steady state side slip angle

velocity
st
(7.108)
at a constant given steering angle
B,
one gets from Eq. (7.108)
st and the steady state angular
= 0
|v| (cS1 + cS2 ) st + (m v |v| + a1 cS1 a2 cS2 ) st = v cS1 0 ,
(7.109)
|v| (a1 cS1 a2 cS2 ) st + (a21 cS1 + a22 cS2 ) st = v a1 cS1 0 ,
(7.110)
where the rst equation has been multiplied by
m |v| |v|
and the second with
|v|.
The solution can be derived from
st =
130
v cS1 0
m v |v| + a1 cS1 a2 cS2
v a1 cS1 0
a21 cS1 + a22 cS2
|v| (cS1 + cS2 )
m v |v| + a1 cS1 a2 cS2
|v| (a1 cS1 a2 cS2 )
a21 cS1 + a22 cS2
(7.111)
and
|v| (cS1 + cS2 )
v cS1 0
|v| (a1 cS1 a2 cS2 ) v a1 cS1 0
st =
(7.112)
|v| (cS1 + cS2 )
m v |v| + a1 cS1 a2 cS2
|v| (a1 cS1 a2 cS2 )
a21 cS1 + a22 cS2
The denominator results in
detD = |v| cS1 cS2 (a1 + a2 )2 + m v |v| (a2 cS2 a1 cS1 )

For a non vanishing denominator
detD 6= 0,
(7.113)
steady state solutions exist
a1
v
cS2 (a1 + a2 )
=
0 ,
|v| a + a + m v |v| a2 cS2 a1 cS1
1
2
cS1 cS2 (a1 + a2 )
a2 m v |v|
st
st =
a1 + a2 + m v |v|
At forward driving vehicles
a2 cS2 a1 cS1 0 .
cS1 cS2 (a1 + a2 )
(7.114)
(7.115)
v > 0, the steady state side slip angle starts with the kinematic
value
v0
st
=
v
a2
0
|v| a1 + a2
v0
st
=
and
v
0
a1 + a2
(7.116)
and decreases with increasing speed. At speeds larger than
s
vst=0 =
a2 cS2 (a1 + a2 )
a1 m
(7.117)
the side slip angle changes the sign. Using the kinematic value of the yaw velocity
Eq. (7.115) can be written as
st =
v
a1 + a2
1
1 +
where
s
vch =
|v|
v
0 ,

(7.118)
v 2
vch

cS1 cS2 (a1 + a2 )2

m (a2 cS2 a1 cS1 )
(7.119)
is called the 'characteristic' speed of the vehicle.
, and the driven curve radius R are plotted versus the

v . The steering angle has been set to 0 = 1.4321 , in order to let the vehicle
In Fig. 7.14 the side slip angle

driving speed
131
Vehicle Dynamics
steady state side slip angle
radius of curvrature
200
0
150
r [m]
[deg]
-2
-4
-6
a1*c S1/a2*c S2 = 0.66667

a1*cS1/a2*c S2 = 1
a1*cS1/a2*c S2 = 1.3333
-8
-10
100
10
20
v [m/s]
m=700 kg;
=1000 kg m2 ;
30
a1 =1.2 m;
a2 =1.3 m;
a1*cS1/a2*cS2 = 0.66667
a1*cS1/a2*cS2 = 1
a1*cS1/a2*cS2 = 1.3333
50
0
40
10
cS1 = 80 000 N m;
20
v [m/s]
cS2
30
40
110 770 N m
= 73 846 N m
55 385 N m
Figure 7.14: Steady state cornering
drive a circle with the radius
R0 = 100 m
at
has been calculated via
st =
v 0.
The actually driven circle radius
v
.
R
(7.120)
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.
7.3.7.2 Steering Tendency

After reaching the steady state solution, the vehicle moves on a circle. When inserting
Eq. (7.120) into Eq. (7.115) and resolving for the steering angle, one gets
0 =
a1 + a2
v 2 v a2 cS2 a1 cS1
+ m
.
R
R |v| cS1 cS2 (a1 + a2 )
(7.121)
The rst term is the Ackermann steering angle, which follows from Eq. (7.2) with the
wheel base
a = a1 + a2
and the approximation for small steering angles
tan 0 0 .
The Ackermann-steering angle provides a good approximation for slowly moving vehicles,
v 0. Depending on
v > 0, backward: v < 0),
because the second expression in Eq. (7.121) becomes very small at

the value of
a2 cS2 a1 cS1
and the driving direction (forward:
the necessary steering angle diers from the Ackermann-steering angle at higher speeds.
The dierence is proportional to the lateral acceleration
ay =
132
v2
.
R
(7.122)
At
v > 0 the
a1 , a2
steering tendency of a vehicle is dened by the position of the center of
gravity
and the cornering stinesses at the axles
cS1 , cS2 .
The various steering
tendencies are arranged in the table 7.1.
understeering
0 > 0A
or
a1 cS1 < a2 cS2
or
a1 cS1
<1
a2 cS2
neutral
0 = 0A
or
a1 cS1 = a2 cS2
or
a1 cS1
=1
a2 cS2
oversteering
0 < 0A
or
a1 cS1 > a2 cS2
or
a1 cS1
>1
a2 cS2
Table 7.1: Steering tendency of a vehicle at forward driving
7.3.7.3 Slip Angles

With the conditions for a steady state solution
st = 0, st = 0
and the relation
Eq. (7.120), the equations of motion Eq. (7.91) and Eq. (7.92) can be dissolved for the
lateral forces
Fy1st =
Fy2st =
a2
v2
,
m
a1 + a2
R
or
a1
v2
m
a1 + a2
R
a1
Fy2st
.
=
a2
Fy1st
(7.123)
With the linear tire model in Eq. (7.74) one gets
st
Fy1
= cS1 sst
y1
where
sst
yA1
and
sst
yA2
and
st
Fy2
= cS2 sst
y2 ,
(7.124)
label the steady state lateral slips at the axles. Now, from Eqs. (7.123)
and (7.124) it follows
st
Fy2
cS2 sst
a1
y2
= st =
a2
Fy1
cS1 sst
y1
or
sst
a1 cS1
y2
= st .
a2 cS2
sy1
That means, at a vehicle with understeering tendency (a1 cS1
< a2 cS2 )
(7.125)
during steady
state cornering the slip angles at the front axle are larger than the slip angles at the rear
st
st
axle, sy1 > sy2 . So, the steering tendency can also be determined from the slip angle at
the axles.
7.3.8 Influence 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,
clin
S1 = cS Fz1
and
clin
S2 = cS Fz2 .
(7.126)
133
Vehicle Dynamics
holds. The weight of the vehicle
G = mg
is distributed over the axles according to the
position of the center of gravity
Fz1 =
a2
G
a1 + a2
.Fz2 =
and
a1
G
a1 + a2
(7.127)
With Eq. (7.126) and Eq. (7.127) one obtains
a1 clin
S1 = a1 cS
a2
G
a1 + a2
(7.128)
a2 clin
S2 = a2 cS
a1
G.
a1 + a2
(7.129)
and
Thus, a vehicle with identical tires would be steering neutrally at a linear inuence of the
wheel load on the cornering stiness, because of
lin
a1 clin
S1 = a2 cS2
(7.130)
The lateral force is applied behind the center of the contact patch at the caster oset
v
distance. Hence, the lever arms of the lateral forces change to a1 a1
n and
|v| L1
v
a2 a2 + |v| nL1 , which will stabilize the vehicle, independently from the driving direction.
6
5
Fz [N ]
Fy [kN]
4
3
2
1
0
Fz [kN]
Figure 7.15: Lateral force
Fy
over wheel load
Fz
Fy [N ]
1000
758
2000
1438
3000
2043
4000
2576
5000
3039
6000
3434
7000
3762
8000
4025
at dierent slip angles
At a real tire, a degressive inuence of the wheel load on the tire forces is observed,
Fig. 7.15. According to Eq. (7.92) the rotation of the vehicle is stable, if the torque from
the lateral forces
Fy1
and
Fy2
is aligning, i.e.
a1 Fy1 a2 Fy2 < 0
(7.131)
a = 2.45 m the axle loads Fz1 = 4000 N and Fz2 =

3000 N yield the position of the center of gravity a1 = 1.05 m and a2 = 1.40 m. At equal
slip on front and rear axle one gets from the table in 7.15 Fy1 = 2576 N and Fy2 = 2043 N .
holds. At a vehicle with the wheel base
134
With this, the condition in Eq. (7.131) yields
1.05 2576 1.45 2043 = 257.55 .
The
value is signicantly negative and thus stabilizing.

Vehicles with
a1 < 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 (a1
> a2 ), generally a stable driving behavior
can only be achieved with dierent tires.

At increasing lateral acceleration the vehicle is more and more supported by the outer
wheels. The wheel load dierences can dier at a suciently rigid vehicle body, because of
dierent kinematics (roll support) or dierent roll stiness. Due to the degressive inuence
of wheel load, the lateral force at an axle decreases with increasing wheel load dierence.
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

2
ay = vR until reaching the limit.
and, with this also the lateral acceleration due
speed
Typical
80
60
side slip angle [deg]
steer angle [deg]
results are displayed in Fig. 8.1.
40
20
0
-2
-4
wheel loads [kN]
roll angle [deg]
5
3
2
1
0
4
3
2
1
0.2
0.4
0.6
0.8
lateral acceleration [g]
0.2
0.4
0.6
0.8
Figure 8.1: Steady state cornering: rear-wheel-driven car on
R = 100 m
The vehicle is under-steering and thus stable according to Eq. (7.121) with Eq. (7.122).
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 dierences between the outer and inner
wheels. With a suciently rigid frame the use of an anti roll bar at the front axle allows
to increase the wheel load dierence there and to decrease it at the rear axle accordingly.
Thus, the digressive inuence of the wheel load on the tire properties, cornering stiness
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 suciently 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.
steering angle [deg]
40
30
20
10
0
0.2
0.4
0.6
time [s]
0.8
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 dier 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
12
0.5
10
0.4
yaw velocity [deg/s]
0.6
0.3
0.2
0.1
6
4
2
2.5
0.5
side slip angle [deg]
roll angle [deg]
Vehicle Dynamics
2
1.5
1
0.5
0
0
-0.5
-1
-1.5
-2
Figure 8.3: Step steer: passenger car at
[t]
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 Profile
The irregularities of a track are of stochastic nature. Fig. 8.4 shows a country road prole in
dierent scalings. To limit the eort 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)
z1 = z1 (s1 ) ,
and
z2 = z2 (s2 )
(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)
138
z = z(s) ;
(8.2)
0.05
0.04
0.03
0.02
0.01
0
-0.01
-0.02
-0.03
-0.04
-0.05
0
10
20
30
40
50
60
70
80
90
100
Figure 8.4: Track irregularities
A normally distributed, stationary and ergodic random process
z = z(s)
is completely
characterized by the rst two expectation values, the mean value
1
mz = lim
s 2s
Zs
z(s) ds
(8.3)
s
and the correlation function
Zs
1
Rzz () = lim
s 2s
z(s) z(s ) ds .
(8.4)
s
A vanishing mean value
mz = 0
can always be achieved by an appropriate coordinate
transformation. The correlation function is symmetric,
Rzz () = Rzz () ,
and
1
Rzz (0) = lim
s 2s
Zs
z(s)
(8.5)
2
ds
(8.6)
s
describes the variance of
zs .
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
Z
Rzz () =
Szz () cos() d
(8.7)
0
where
denotes the space circular frequency. With Eq. (8.7) follows from Eq. (8.6)
Z
Rzz (0) =
Szz () d .
(8.8)
139
Vehicle Dynamics
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

Szz () = S0
w
,
(8.9)
0 = 1 m1 . The reference psd S0 = Szz (0 )

acts as a measurement for unevennes and the waviness w indicates, whether the track has
where the reference frequency is xed to
notable irregularities in the short or long wave spectrum. At real tracks, the reference6
psd S0 lies within the range from 1 10
m3 to 100 106 m3 and the waviness can be
approximated by
w = 2.
8.1.3.2 Steering Activity

-6
highway: S 0=1*10
-5
m ; w=2
country road: S0=2*10
1000
1000
500
500
-2
[deg] 2
-2
m ; w=2
[deg] 2
Figure 8.5: Steering activity on dierent 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 = 90 km/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 dierence 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.
140
vehicle
mass
[kg]
center of gravity
[m]
empty
12 500
3.800 | 0.000 | 1.500
fully laden
18 000
3.860 | 0.000 | 1.600
inertias
[kg m2 ]
12 500
0
0
0
155 000
0
0
0
155 000
15 400
0
250
0
200 550
0
250
0
202 160
Table 8.1: Data for a laden and empty coach
The coach has a wheel base of
a = 6.25m. The front axle with the track width sv = 2.046m
has a double wishbone single wheel suspension. The twin-tire rear axle with the track
i
o
widths sh = 2.152 m and sh = 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.
suspension travel [cm]
8.2.2 Roll Steering

10
5
0
-5
-10
-1
0
steer angle [deg]
Figure 8.6: Roll steer: - - front, rear

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 eect,
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 occu-
pied 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.
141
Vehicle Dynamics
steer angle
250
LW
[deg]
vehicle course
200
200
[m]
150
150
100
100
50
50
0.1 0.2 0.3 0.4

lateral acceleration a y [g]
wheel loads [kN]
100
50
-100
0
[m]
100
wheel loads [kN]
100
50
0.1 0.2 0.3 0.4

0.1 0.2 0.3 0.4

Figure 8.7: Steady state cornering: coach - - empty,
fully occupied
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 o at a lateral acceleration of
ay 0.4 g
. If the vehicle is fully occupied, this eect will occur already at
ay 0.35 g .
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 stiness increases with the load.
Because the damper eect 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.
142
yaw velocity

0.4
10
0.2
0.1
2
0
roll angle
[deg]
side slip angle
[deg]
0
-1
2
0
Z [deg/s]
0.3
-2
0
[s] 6
Figure 8.8: Step steer input:
- - coach empty,
[s] 6
coach fully occupied
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, .
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 dierence at the rear axle, Fig. 8.10. Because the wheel
load dierence is simultaneously reduced at the front axle, the understeering tendency is
reduced. In the limit range, this even leads to an oversteering behavior.
143
Vehicle Dynamics
vertical motion [cm]
10
Figure 8.9: Rear axle:
steer angle
5
0
-5
-10
-5
semi-trailing arm, - - single wishbone,
LW
100
0
lateral motion [cm]
[deg]
roll angle
trailing arm
[Grad]
4
3
50
2
1
0.2
0.4
0.6
0.8
wheel loads front [kN]
2
0
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8
wheel loads rear [kN]
0.2
0.4
0.6
0.8
Figure 8.10: Steady state cornering, semi-trailing arm, - - single wishbone,
trailing
arm
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.
144
Index
Ackermann geometry, 109
Damping rate, 76
Ackermann steering angle, 109, 132
Deviation, 13
Aerodynamic forces, 96
Disturbance-reaction problem, 84
Air resistance, 96
Disturbing force lever, 9
All wheel drive, 118
Down forces, 96
Anti dive, 108
Downhill capacity, 97
Anti roll bar, 122
Drag link, 57, 58
Anti squat, 108
Drive pitch angle, 103
Anti-lock-system, 102
Driver, 3
Auto-correlation, 13
Driving force distribution, 100
Axle kinematics, 108
Driving safety, 72
Double wishbone, 7
Dynamic axle load, 95
McPherson, 7
Dynamic force elements, 63
Multi-link, 7
Dynamic wheel loads, 94
Axle load, 95
Axle suspension
Solid axle, 55
Twist beam, 56
Bend angle, 113, 116
Brake pitch angle, 103
Brake pitch pole, 108
Braking force distribution, 100
Camber angle, 6, 24
Camber compensation, 121, 124
Camber slip, 49
Caster, 8, 9
Characteristic speed, 131
Eective value, 13
Eigenvalues, 128
Environment, 4
First harmonic oscillation, 63
Fourier-approximation, 64
Frequency domain, 63
Friction, 97
Front wheel drive, 98, 118
Generalized uid mass, 70
Grade, 95
Hydro-mount, 69
Climbing capacity, 97
Kingpin, 7
Comfort, 72
Kingpin Angle, 8
Contact point, 24
Cornering resistance, 117, 118
Lateral acceleration, 121, 132
Cornering stiness, 41, 133
Lateral force, 126
Curvature gradient, 114
Lateral slip, 126
Vehicle Dynamics
Ljapunov equation, 84
State vector, 88
Load, 4
Steady state cornering, 117, 136, 141
Maximum acceleration, 97, 98

Maximum deceleration, 97, 99
Mean value, 13
Natural frequency, 76
Steering activity, 140

Steering angle, 114
Steering box, 57, 58
Steering lever, 58
Steering oset, 9
Steering system
Optimal damping, 81, 87
Drag link steering system, 58
Oversteering, 133
Lever arm, 57
Overturning limit, 118
Rack and pinion, 57
Parallel track model, 10

Parallel tracks, 138
Pinion, 57
Pivot pole, 109
Power spectral density, 139
Quarter car model, 87, 90
Steering tendency, 125, 132

Step steer input, 137, 143
Suspension model, 75
Suspension spring rate, 78
System response, 63
Tilting condition, 97
Tire
Rack, 57
Bore slip, 52
Random road prole, 138
Bore torque, 20, 50
Rear wheel drive, 98, 118
Camber angle, 24
Reference frames
Camber inuence, 48
Ground xed, 5
Characteristics, 52
Inertial, 5
Circumferential direction, 24
Vehicle xed, 5
Composites, 19
Relative damping rate, 77
Contact forces, 20
Ride comfort, 83
Contact patch, 20
Ride safety, 83
Contact point, 23
Road, 10, 23
Contact point velocity, 31
Roll axis, 124
Contact torques, 20
Roll center, 124
Cornering stiness, 41
Roll steer, 141
Deection, 26
Roll stiness, 120
Deformation velocity, 31
Roll support, 121, 124
Development, 19
Rolling condition, 126
Dynamic oset, 41
Dynamic radius, 32, 33
Safety, 72
Friction coecient, 45
Side slip angle, 109
Lateral direction, 24
Sky hook damper, 87
Lateral force, 20
Space requirement, 110
Lateral force characteristics, 41
Spring rate, 78
Lateral force distribution, 40
Stability, 128
Lateral slip, 40
State equation, 128
Lateral velocity, 31
State matrix, 88
ii
Lift o, 88
Vehicle model, 75, 90, 94, 103, 112, 121,
Linear model, 126
125
Loaded radius, 24, 32
Vertical dynamics, 72
Longitudinal force, 20, 38, 39
Virtual work, 121
Longitudinal force characteristics, 39

Longitudinal force distribution, 39
Longitudinal slip, 39
Longitudinal velocity, 31
Model, 52
Normal force, 20
Pneumatic trail, 41
Radial damping, 35
Radial direction, 24
Radial stiness, 34, 121
Rolling resistance, 20, 36, 37
Self aligning torque, 20, 41
Sliding velocity, 40
Static radius, 24, 32, 33
Tilting torque, 20
Track normal, 24, 26
Transport velocity, 33
Tread deection, 38
Tread particles, 37
Unloaded radius, 32
Vertical force, 34
Wheel load inuence, 41
Waviness, 140
Wheel
Angular velocity, 50
Wheel base, 109
Wheel load, 20
Wheel loads, 94
Wheel rotation axis, 5
Wheel Suspension
Semi-trailing arm, 143
Single wishbone, 143
Trailing arm, 143
Wheel suspension
Central control arm, 56
Double wishbone, 55
McPherson, 55
Multi-Link, 55
Semi-trailing arm, 56
SLA, 56
Yaw angle, 112, 115
Yaw velocity, 126
Tire Model
Kinematic, 109
Linear, 133
TMeasy, 52
Toe angle, 6
Toe-in angle, 6
Track, 23
Track curvature, 114
Track normal, 5
Track radius, 114
Track width, 109, 121
Trailer, 112, 115
Understeering, 133
Variance, 13
Vehicle, 3
Vehicle comfort, 72
Vehicle dynamics, 2
iii

Vehicle Dynamics - Rill PDF

Uploaded by

Copyright:

Available Formats

You might also like

Vehicle Dynamics - Rill PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Vehicle Dynamics - Rill PDF

Uploaded by

Copyright:

Available Formats

Toe and camber angle

Design Position of Wheel Rotation Axis

Caster and Kingpin Angle . . . . . . . . . . . . . . . . . .

Caster, Steering Oset and Disturbing Force Lever

Bumps and Potholes

Classication of Random Road Proles . . . . . . . . . . . . . . . .

Tire Forces and Torques

Measuring Tire Forces and Torques . . . . . . . . . . . . . . . . . .

Local Track Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Length of Contact Patch . . . . . . . . . . . . . . . . . . . . . . . .

Static Contact Point

Contact Point Velocity . . . . . . . . . . . . . . . . . . . . . . . . .

Dynamic Rolling Radius

Forces and Torques caused by Pressure Distribution . . . . . . . . . . . . .

Friction Forces and Torques

Longitudinal Force and Longitudinal Slip . . . . . . . . . . . . . . .

Lateral Slip, Lateral Force and Self Aligning Torque . . . . . . . . .

Wheel Load Inuence . . . . . . . . . . . . . . . . . . . . . . . . . .

Two-Dimensional Tire Characteristics . . . . . . . . . . . . . . . . .

Dierent Friction Coecients

Self Aligning Torque

Typical Tire Characteristics

Purpose and Components

Multi Purpose Systems . . . . . . . . . . . . . . . . . . . . . . . . .

Rack and Pinion Steering

Lever Arm Steering System

Bus Steer System . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Drag Link Steering System . . . . . . . . . . . . . . . . . . . . . . .

Standard Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Testing and Evaluating Procedures

Simple Spring Damper Combination

General Dynamic Force Model . . . . . . . . . . . . . . . . . . . . .

Dynamic Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Full Vehicle Model

Natural Frequency and Damping Rate

Minimum Spring Rates . . . . . . . . . . . . . . . . . . . .

Fast Approach to Steady State

Sky Hook Damper

Eigenfrequencies and Damping Ratios . . . . . . . . . . . . . . . . .

Nonlinear Force Elements

Quarter Car Model . . . . . . . . . . . . . . . . . . . . . . . . . . .

Dynamic Wheel Loads

Simple Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . .

Single Axle Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Braking at Single Axle . . . . . . . . . . . . . . . . . . . . . . . . .

Optimal Distribution of Drive and Brake Forces . . . . . . . . . . . 100

Dierent Distributions of Brake Forces

Driving and Braking

Drive and Brake Pitch

Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Driving and Braking

Brake Pitch Pole

Kinematic Tire Model

Space Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

Vehicle Model with Trailer . . . . . . . . . . . . . . . . . . . . . . . 112

Entering a Curve . . . . . . . . . . . . . . . . . . . . . . . 114

Trailer Motions . . . . . . . . . . . . . . . . . . . . . . . . 115

Caster, Steering Oset and Disturbing Force Lever

Classication of Random Road Proles . . . . . . . . . . . . . . . .

Wheel Load Inuence . . . . . . . . . . . . . . . . . . . . . . . . . .

Dierent Friction Coecients

Dierent Distributions of Brake Forces

Inuence of Wheel Load on Cornering Stiness

Random Road Prole

Coach with dierent Loading Conditions

Dierent Rear Axle Concepts for a Passenger Car

H.: Dynamik der Kraftfahrzeuge. 4. Auage.

The environment also inuences the driver:

irregularities, coecient of friction

but also aects the driver:

x0 , y0 , z0 serves as an inertial reference frame. Within the vehicle-xed

is xed to the wheel carrier. In design position its axes

the corresponding axis of vehicle-xed reference frame