VEHICLE DYNAMICS

FACHHOCHSCHULE REGENSBURG
UNIVERSITY OF APPLIED SCIENCES
HOCHSCHULE FR
TECHNIK
WIRTSCHAFT
SOZIALES

LECTURE NOTES
Prof. Dr. Georg Rill
October 2003

download: http://homepages.fh-regensburg.de/%7Erig39165/
Contents

Contents I

1 Introduction 1
1.1 Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Vehicle Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Driver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.3 Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.4 Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.5 Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Wheel/Axle Suspension Systems . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.1 General Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.2 Multi Purpose Suspension Systems . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Specific Suspension Systems . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Steering Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.2 Rack and Pinion Steering . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.3 Lever Arm Steering System . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.4 Drag Link Steering System . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.5 Bus Steer System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.1 Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.2 Toe and Camber Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4.2.1 Definitions according to DIN 70 000 . . . . . . . . . . . . . . . 9
1.4.2.2 Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4.3 Steering Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4.3.1 Kingpin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4.3.2 Caster and Kingpin Angle . . . . . . . . . . . . . . . . . . . . 11
1.4.3.3 Disturbing Force Lever, Caster and Kingpin Offset . . . . . . . 12

2 The Tire 13
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.1 Tire Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.2 Tire Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.3 Forces and Torques in the Tire Contact Area . . . . . . . . . . . . . . . 14
2.2 Contact Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

I
 2.2.1 Dynamic Rolling Radius . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2.2 Contact Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.3 Local Track Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.4 Contact Point Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3 Wheel Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Longitudinal Force and Longitudinal Slip . . . . . . . . . . . . . . . . . . . . . 20
2.5 Lateral Slip, Lateral Force and Self Aligning Torque . . . . . . . . . . . . . . . . 23
2.6 Camber Influence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.7 Bore Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.8 Typical Tire Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3 Longitudinal Dynamics 30
3.1 Dynamic Wheel Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.1.1 Simple Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.1.2 Influence of Grade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.1.3 Aerodynamic Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2 Maximum Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.1 Tilting Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.2 Friction Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3 Driving and Braking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.3.1 Single Axle Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.3.2 Braking at Single Axle . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.3.3 Optimal Distribution of Drive and Brake Forces . . . . . . . . . . . . . . 36
3.3.4 Different Distributions of Brake Forces . . . . . . . . . . . . . . . . . . . 38
3.3.5 Anti-Lock-Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.4 Drive and Brake Pitch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.4.1 Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.4.2 Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4.3 Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4.4 Driving and Braking . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.5 Brake Pitch Pole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

4 Lateral Dynamics 45
4.1 Kinematic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.1 Kinematic Tire Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.2 Ackermann Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.3 Space Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.1.4 Vehicle Model with Trailer . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.1.4.1 Position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.1.4.2 Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1.4.3 Entering a Curve . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.1.4.4 Trailer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.1.4.5 Course Calculations . . . . . . . . . . . . . . . . . . . . . . . 52
4.2 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

II
 4.2.1 Overturning Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2.2 Roll Support and Camber Compensation . . . . . . . . . . . . . . . . . 56
4.2.3 Roll Center and Roll Axis . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.2.4 Wheel Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2.5 Cornering Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.3 Simple Handling Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3.2 Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3.3 Lateral Slips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.3.4 Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.3.5 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.5.1 Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.5.2 Low Speed Approximation . . . . . . . . . . . . . . . . . . . . 66
4.3.5.3 High Speed Approximation . . . . . . . . . . . . . . . . . . . . 66
4.3.6 Steady State Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.3.6.1 Side Slip Angle and Yaw Velocity . . . . . . . . . . . . . . . . 67
4.3.6.2 Steering Tendency . . . . . . . . . . . . . . . . . . . . . . . . 69
4.3.6.3 Slip Angles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3.7 Influence of Wheel Load on Cornering Stiffness . . . . . . . . . . . . . 70

5 Vertical Dynamics 73
5.1 Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2 Basic Tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.1 Simple Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.2 Track . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2.3 Spring Preload . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2.4 Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.5 Free Vibrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3 Nonlinear Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.3.1 Quarter Car Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.3.2 Random Road Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.3.3 Vehicle Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.3.4 Quality Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.3.5 Optimal Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.3.5.1 Linear Characteristics . . . . . . . . . . . . . . . . . . . . . . 81
5.3.5.2 Nonlinear Characteristics . . . . . . . . . . . . . . . . . . . . 81
5.3.5.3 Limited Spring Travel . . . . . . . . . . . . . . . . . . . . . . . 83
5.4 Dynamic Force Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.4.1 System Response in the Frequency Domain . . . . . . . . . . . . . . . 84
5.4.1.1 First Harmonic Oscillation . . . . . . . . . . . . . . . . . . . . 84
5.4.1.2 Sweep-Sine Excitation . . . . . . . . . . . . . . . . . . . . . . 86
5.4.2 Hydro-Mount . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.4.2.1 Principle and Model . . . . . . . . . . . . . . . . . . . . . . . 87
5.4.2.2 Dynamic Force Characteristics . . . . . . . . . . . . . . . . . 89

III
 5.5 Different Influences on Comfort and Safety . . . . . . . . . . . . . . . . . . . . 90
5.5.1 Vehicle Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.5.2 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

6 Driving Behavior of Single Vehicles 93

6.1 Standard Driving Maneuvers . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.1.1 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.1.2 Step Steer Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.1.3 Driving Straight Ahead . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.1.3.1 Random Road Profile . . . . . . . . . . . . . . . . . . . . . . 95
6.1.3.2 Steering Activity . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.2 Coach with different Loading Conditions . . . . . . . . . . . . . . . . . . . . . 98
6.2.1 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.2.2 Roll Steer Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.2.3 Steady State Cornering . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.2.4 Step Steer Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.3 Different Rear Axle Concepts for a Passenger Car . . . . . . . . . . . . . . . . 100

IV
1 Introduction

1.1 Terminology

1.1.1 Vehicle Dynamics

The Expression Vehicle Dynamics encompasses the interaction of

driver,

vehicle

load and

environment

Vehicle dynamics mainly deals with

the improvement of active safety and driving comfort as well as

the reduction of road destruction.

In vehicle dynamics

computer calculations

test rig measurements and

field tests

are employed.
The interactions between the single systems and the problems with computer calculations
and/or measurements shall be discussed in the following.

1
Vehicle Dynamics FH Regensburg, University of Applied Sciences

1.1.2 Driver

By various means of interference the driver can interfere with the vehicle:


steering wheel
lateral dynamics

gas pedal




driver brake pedal vehicle

longitudinal dynamics
clutch






gear shift

The vehicle provides the driver with some information:

vibrations: longitudinal, lateral, vertical
vehicle sound: motor, aerodynamics, tires driver
instruments: velocity, external temperature, ...

The environment also influences the driver:

climate
environment traffic density driver
track

A drivers reaction is very complex. To achieve objective results, an ideal driver is used in
computer simulations and in driving experiments automated drivers (e.g. steering machines)
are employed.
Transferring results to normal drivers is often difficult, if field tests are made with test drivers.
Field tests with normal drivers have to be evaluated statistically. In all tests, the drivers security
must have absolute priority.
Driving simulators provide an excellent means of analyzing the behavior of drivers even in limit
situations without danger.
For some years it has been tried to analyze the interaction between driver and vehicle with
complex driver models.

1.1.3 Vehicle

The following vehicles are listed in the ISO 3833 directive:

Motorcycles,

Passenger Cars,

Busses,

Trucks

2
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

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 motions and 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 field and/or laboratory
tests.
Results can be falsified by safety devices, e.g. jockey wheels on trucks.

1.1.4 Load

Trucks are conceived for taking up load. Thus their driving behavior changes.

mass, inertia, center of gravity
Load
dynamic behaviour (liquid load)
In computer calculations problems occur with the determination of the inertias and the mod-
elling of liquid loads.
Even the loading and unloading process of experimental vehicles takes some effort. When
making experiments with tank trucks, flammable liquids have to be substituted with water. The
results thus achieved cannot be simply transferred to real loads.

1.1.5 Environment

The Environment influences primarily the vehicle:

 
Road: irregularities, coefficient of friction
Environment vehicle
Air: resistance, cross wind
but also influences the driver
 
climate
Environment driver
visibility
Through the interactions between vehicle and road, roads can quickly be destroyed.
The greatest problem in field test and laboratory experiments is the virtual impossibility of
reproducing environmental influences.
The main problems in computer simulation are the description of random road irregularities
and the interaction of tires and road as well as the calculation of aerodynamic forces and
torques.

3
Vehicle Dynamics FH Regensburg, University of Applied Sciences

1.2 Wheel/Axle Suspension Systems

1.2.1 General Remarks

The Automotive Industry uses different kinds of wheel/axle suspension systems. Important
criteria are costs, space requirements, kinematic properties and compliance attributes.

1.2.2 Multi Purpose Suspension Systems

The Double Wishbone Suspension, the McPherson Suspension and the Multi-Link Suspension
are multi purpose wheel suspension systems, Fig. 1.1.

E
E
G
zR O G
zR
zR
2
xR
yR O2 F D F
D N3 yR Y
yR P
xR O1 S Z
Q xR W
R G P X
S Q 1 U R P
R Q
zB zB S
N1 F B D V
U2 A
B U
S
xB xB
U1 yB yB
C
M M A

Figure 1.1: Double Wishbone, McPherson and Multi-Link Suspension

They are used as steered front or non steered rear axle suspension systems. These suspen-
sion 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.
At pickups, trucks and busses often solid axles are used. Solid axles are guided either by leaf

zA
zA Z2
Y2
Z1
Y1

X2
xA xA
X1
yA yA

Figure 1.2: Solid Axles

4
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

springs or by rigid links, Fig. 1.2. Solid axles tend to tramp on rough road.
Leaf spring guided solid axle suspension systems are very robust. Dry friction between the
leafs leads to locking effects in the suspension. Although the leaf springs provide axle guidance
on some solid axle suspension systems additional links in longitudinal and lateral direction are
used. Thus the typical wind up effect on braking can be avoided.
Solid axles suspended by air springs need at least four links for guidance. In addition to a good
driving comfort air springs allow level control too.

1.2.3 Specific Suspension Systems

The Semi-Trailing Arm, the SLA and the Twist Beam axle suspension are suitable only for non
steered axles, Fig. 1.3.

zR
zA

yR yA
xR

xA

Figure 1.3: Specific 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 SLA axle design allows a nearly independent layout of longitudinal and lateral axle mo-
tions. It is similar to the Central Control Arm axle suspension, where the trailing arm is com-
pletely rigid and hence only two lateral links are needed.
The twist beam axle suspension exhibits either a trailing arm or a semi-trailing arm character-
istic. It is used for non driven rear axles only. The twist beam axle provides enough space for
spare tire and fuel tank.

1.3 Steering Systems

1.3.1 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.

5
Vehicle Dynamics FH Regensburg, University of Applied Sciences

In order to achieve a good maneuverability a maximum steer angle of approx. 30 must be

provided at the front wheels of passenger cars. Depending on the wheel base busses and
trucks need maximum steer angles up to 55 at the front wheels.
Recently some companies have started investigations on steer by wire techniques.

1.3.2 Rack and Pinion Steering

Rack and pinion is the most common steering system on passenger cars, Fig. 1.4. The rack
may be located either in front of or behind the axle. The rotations of the steering wheel L are

uZ L
P pinion
nk
drag li
Q rack steer
wheel
1 box 2
and
wheel
body L

Figure 1.4: Rack and Pinion Steering

firstly transformed by the steering box to the rack travel uZ = uZ (L ) and then via the drag
links transmitted to the wheel rotations 1 = 1 (uZ ), 2 = 2 (uZ ). Hence the overall steering
ratio depends on the ratio of the steer box and on the kinematics of the steer linkage.

1.3.3 Lever Arm Steering System

Using a lever arm steering system Fig. 1.5, large steer angles at the wheels are possible. This

G
steer box
ste
e r le 2
ver r le ver
Q1 1 stee Q2

drag link 1 P1 P2 drag link 2

1 L
2
wheel and
wheel body

Figure 1.5: Lever Arm Steering System

6
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

steering system is used on trucks with large wheel bases and independent wheel suspension
at the front axle. Here the steering box can be placed outside of the axle center.
The rotations of the steering wheel L are firstly transformed by the steering box to the ro-
tation of the steer levers G = G (L ). The drag links transmit this rotation to the wheel
1 = 1 (G ), 2 = 2 (G ). Hence, again the overall steering ratio depends on the ratio of
the steer box and on the kinematics of the steer linkage.

1.3.4 Drag Link Steering System

At solid axles the drag link steering system is used, Fig. 1.6.

ver
steer le steer box
H (90o rotated)
O
H

wheel
and steer link
wheel
body

I
L
1 2

K drag link

Figure 1.6: Drag Link Steering System

The rotations of the steering wheel L are transformed by the steering box to the rotation of the
steer lever arm H = H (L ) and further on to the rotation of the left wheel, 1 = 1 (H ). The
drag link transmits the rotation of the left wheel to the right wheel, 2 = 2 (1 ). The steering
ratio is defined by the ratio of the steer box and the kinematics of the steer link. Here the ratio
2 = 2 (1 ) given by the kinematics of the drag link can be changed separately.

1.3.5 Bus Steer System

In busses the driver sits more than 2 m in front of the front axle. Here, sophisticated steer
systems are needed, Fig. 1.7.
The rotations of the steering wheel L are transformed by the steering box to the rotation of
the steer lever arm H = H (L ). Via the steer link the left lever arm is moved, H = H (G ).
This motion is transferred by a coupling link to the right lever arm. Via the drag links the left
and right wheel are rotated, 1 = 1 (H ) and 2 = 2 (H ).

7
Vehicle Dynamics FH Regensburg, University of Applied Sciences

steer
lever
G
H
steer box
steer link

left
lever
arm J
I
K
Q drag link H coupl.
link
P
1 L
2
wheel and
wheel body

Figure 1.7: Bus Steer System

1.4 Definitions

1.4.1 Coordinate Systems

In vehicle dynamics several different coordinate systems are used, Fig 1.8.

z0
y0
x0

zF
yF
xF

en eyR
ey ex

Figure 1.8: Coordinate Systems

The inertial system with the axes x0 , y0 , z0 is fixed to the track. Within the vehicle fixed system
the xF -axis is pointing forward, the yF -axis left and the zF -axis upward. The orientation of the
wheel is given by the unit vector eyR in direction of the wheel rotation axis.

8
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

The unit vectors in the directions of circumferential and lateral forces ex and ey as well as the
track normal en follow from the contact geometry.

1.4.2 Toe and Camber Angle

1.4.2.1 Definitions according to DIN 70 000

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 angle. It is positive, if the front part of
the wheel is oriented towards the vehicle center plane, Fig. 1.9.

front

xF

yF
left right
rear

Figure 1.9: Positive Toe Angle

The camber angle is the angle between the wheel center plane and the track normal. It is
positive, if the upper part of the wheel is inclined outwards, Fig. 1.10.

top
zF

left yF
right
bottom

Figure 1.10: Positive Camber Angle

1.4.2.2 Calculation

The calculation of the toe angle is done for the left wheel. The unit vector eyR in direction of
the wheel rotation axis is described in the vehicle fixed coordinate system F , Fig. 1.11
h iT
(1) (2) (3)
eyR,F = eyR,F eyR,F eyR,F , (1.1)

where the axis xF and zF span the vehicle center plane. The xF -axis points forward and the
zF -axis points upward. The toe angle can then be calculated from

9
Vehicle Dynamics FH Regensburg, University of Applied Sciences

eyR

zF

yF xF
e (2) V e (3)
yR,F yR,F

e (1)
yR,F

Figure 1.11: Toe Angle

(1)
eyR,F
tan = (2)
. (1.2)
eyR,F

The real camber angle follows from the scalar product between the unit vectors in the direc-
tion of the wheel rotation axis eyR and in the direction of the track normal en ,

sin = eTn eyR . (1.3)

The wheel camber angle can be calculated by

(3)
sin = eyR,F . (1.4)

On a flat horizontal road both definitions are equal.

1.4.3 Steering Geometry

1.4.3.1 Kingpin

At the steered front axle the McPherson-damper strut axis, the double wishbone axis and
multi-link wheel suspension or dissolved double wishbone axis are frequently employed in
passenger cars, Fig. 1.12 and Fig. 1.13.
The wheel body rotates around the kingpin at steering movements.
At the double wishbone axis, the ball joints A and B , which determine the kingpin, are fixed to
the wheel body.
The ball joint point A is also fixed to the wheel body at the classic McPherson wheel suspen-
sion, but the point B is fixed to the vehicle body.
At a multi-link axle, the kingpin is no longer defined by real link points. Here, as well as with
the McPherson wheel suspension, the kingpin changes its position against the wheel body at
wheel travel and steer motions.

10
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

zR

B
yR
M xR

kingpin axis A-B

Figure 1.12: Double Wishbone Wheel Suspension

B
zR zR

yR yR
xR M xR
M
A

kingpin axis A-B rotation axis

Figure 1.13: McPherson and Multi-Link Wheel Suspensions

1.4.3.2 Caster and Kingpin Angle

The current direction of the kingpin can be defined by two angles within the vehicle fixed
coordinate system, Fig. 1.14.
If the kingpin is projected into the yF -, zF -plane, the kingpin inclination angle can be read as
the angle between the zF -axis and the projection of the kingpin.
The projection of the kingpin into the xF -, zF -plane delivers the caster angle with the angle
between the zF -axis and the projection of the kingpin.
With many axles the kingpin and caster angle can no longer be determined directly.
The current rotation axis at steering movements, that can be taken from kinematic calculations
here delivers a virtual kingpin. The current values of the caster angle and the kingpin incli-
nation angle can be calculated from the components of the unit vector in the direction of the

11
Vehicle Dynamics FH Regensburg, University of Applied Sciences

zF
zF eS

yF
xF

Figure 1.14: Kingpin and Caster Angle

kingpin, described in the vehicle fixed coordinate system

(1) (2)
eS,F eS,F
tan = (3)
and tan = (3)
(1.5)
eS,F eS,F
with h iT
(1) (2) (3)
eS,F = eS,F eS,F eS,F . (1.6)

1.4.3.3 Disturbing Force Lever, Caster and Kingpin Offset

The distance d between the wheel center and the king pin axis is called disturbing force lever.
It is an important quantity in evaluating the overall steer behavior.
In general, the point S where the kingpin runs through the track plane does not coincide with
the contact point P , Fig. 1.15.

C d

ey ex
P
S

rS nK

Figure 1.15: Caster and Kingpin Offset

If the kingpin penetrates the track plane before the contact point, the kinematic kingpin offset
is positive, nK > 0.
The caster offset is positive, rS > 0, if the contact point P lies outwards of S .

12
2 The Tire

2.1 Introduction

2.1.1 Tire Development

The following table shows some important mile stones in the development of tires.

1839 Charles Goodyear: vulcanization

1845 Robert William Thompson: first pneumatic tire
(several thin inflated 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 deflations,
22 complete inner tube changes
1899 Continental: longer life 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 2.1: Mile Stones in the Development of Tires

2.1.2 Tire Composites

A modern tire is a mixture of steel, fabric, and rubber.

13
Vehicle Dynamics FH Regensburg, University of Applied Sciences

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%

Tire Mass 8.5 kg

Table 2.2: Tire Composites: 195/65 R 15 ContiEcoContact, Data from www.felge.de

2.1.3 Forces and Torques in the Tire Contact Area

In any point of contact between tire and track normal and friction forces are delivered. Accord-
ing to the tires profile design the contact area forms a not necessarily coherent area.
The effect of the contact forces can be fully described by a vector of force and a torque in
reference to a point in the contact patch. The vectors are described in a track-fixed coordinate
system. The z -axis is normal to the track, the x-axis is perpendicular to the z -axis and perpen-
dicular to the wheel rotation axis eyR . The demand for a right-handed coordinate system then
also fixes the y -axis.

Fx longitudinal or circumferential force

Fy lateral force
Fz vertical force or wheel load

Mx tilting torque Fy 

My rolling resistance torque

My
Mx 

Mz

Mz self aligning and bore torque Fx Fz

Figure 2.1: Contact Forces and Torques

The components of the contact force are named according to the direction of the axes, Fig. 2.1.
Non symmetric distributions of force in the contact patch cause torques around the x and y
axes. The tilting torque Mx occurs when the tire is cambered. My also contains the rolling
resistance of the tire. In particular the torque around the z -axis is relevant in vehicle dynamics.
It consists of two parts,
Mz = M B + M S . (2.1)
Rotation of the tire around the z -axis causes the bore torque MB . The self aligning torque MS
respects the fact that in general the resulting lateral force is not applied in the center of the
contact patch.

14
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

2.2 Contact Geometry

2.2.1 Dynamic Rolling Radius

At an angular rotation of 4, assuming the tread particles stick to the track, the deflected tire
moves on a distance of x, Fig. 2.2.

deflected tire rigid wheel

r0 r rD
S
vt

x x

Figure 2.2: Dynamic Rolling Radius

With r0 as unloaded and rS = r0 4r as loaded or static tire radius

r0 sin 4 = x (2.2)

and
r0 cos 4 = rS . (2.3)
hold.
If the movement of a tire is compared to the rolling of a rigid wheel, its radius rD then has to
be chosen so, that at an angular rotation of 4 the tire moves the distance

r0 sin 4 = x = rD 4 . (2.4)

Hence, the dynamic tire radius is given by

r0 sin 4
rD = . (2.5)
4
For 4 0 one gets the trivial solution rD = r0 .
At small, yet finite angular rotations the sine-function can be approximated by the first terms of
its Taylor-Expansion. Then, (2.5) reads as

4 61 43
 
1
rD = r0 = r0 1 42 . (2.6)
4 6

15
Vehicle Dynamics FH Regensburg, University of Applied Sciences

With the according approximation for the cosine-function

 
rS 1 rS
= cos 4 = 1 42 or 42 = 2 1 (2.7)
r0 2 r0

one finally gets   

1 rS 2 1
rD = r0 1 1 = r0 + rS (2.8)
3 r0 3 3
remains.
The radius rD depends on the wheel load Fz because of rS = rS (Fz ) and thus is named
dynamic tire radius. With this first approximation it can be calculated from the undeformed
radius r0 and the steady state radius rS .
By
v t = rD (2.9)
the average velocity is given with which tread particles are transported through the contact
area.

2.2.2 Contact Point

The current position of a wheel in relation to the fixed x0 -, y0 - z0 -system is given by the wheel
center M and the unit vector eyR in the direction of the wheel rotation axis, Fig. 2.3.

tire rim centre plane

ezR

M
M e yR
e yR
en
rS
en ex
b
P0 a
P0 z0 P
ey
x0
P*
y0
road: z = z ( x , y ) 0 local road plane

Figure 2.3: Contact Geometry

The irregularities of the track can be described by an arbitrary function of two spatial coordi-
nates
z = z(x, y). (2.10)

16
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

At an uneven track the contact point P can not be calculated directly. One can firstly get an
estimated value with the vector
rM P = r0 ezB , (2.11)
where r0 is the undeformed tire radius and ezB is the unit vector in the z -direction of the body
fixed reference frame.
The position of P with respect to the fixed system x0 , y0 , z0 is determined by

r0P = r0M + rM P , (2.12)

where the vector r0M states the position of the rim center M . Usually the point P lies not on
the track. The corresponding track point P0 follows from
(

r0P ,0 1)
(
r0P0 ,0 = r0P ,0 2) . (2.13)

( (
z(r0P ,0 1), r0P ,0 2))

In the point P0 now the track normal en is calculated. Then the unit vectors in the tires circum-
ferential direction and lateral direction can be calculated
eyR en
ex = , and ey = en ex . (2.14)
| eyR en |
Calculating ex demands a normalization, for the unit vector in the direction of the wheel rotation
axis eyR is not always perpendicular to the track. The tire camber angle

= arcsin eTyR en


(2.15)

describes the inclination of the wheel rotation axis against the track normal.
The vector from the rim center M to the track point P0 is now split into three parts

rM P0 = rS ezR + a ex + b ey , (2.16)

where rS names the loaded or static tire radius and a, b are displacements in circumferential
and lateral direction.
The unit vector
ex eyR
ezR = . (2.17)
| ex eyR |
is perpendicular to ex and eyR . Because the unit vectors ex and ey are perpendicular to en , the
scalar multiplication of (2.16) with en results in

eTn rM P0
eTn rM P0 = rS eTn ezR or rS = . (2.18)
eTn ezR

Now also the tire deflection can be calculated

4r = r0 rS , (2.19)

17
Vehicle Dynamics FH Regensburg, University of Applied Sciences

with r0 marking the undeformed tire radius.

The point P given by the vector
rM P = rS ezR (2.20)
lies within the rim center plane. The transition from P 0 to P takes place according to (2.16)
by terms a ex and b ey , standing perpendicular to the track normal. The track normal however
was calculated in the point P 0 . Therefore with an uneven track P no longer lies on the track.
With the newly estimated value P = P now the equations (2.13) to (2.20) can be recurred
until the difference between P and P0 is sufficiently small.
Tire models which can be simulated within acceptable time assume that the contact patch is
even. At an ordinary passenger-car tire, the contact patch has at normal load about the size of
approximately 2020 cm. There is obviously little sense in calculating a fictitious contact point
to fractions of millimeters, when later the real track is approximated in the range of centimeters
by a plane.
If the track in the contact patch is replaced by a plane, no further iterative improvement is
necessary at the hereby used initial value.

2.2.3 Local Track Plane

A plane is given by three points. With the tire width b, the undeformed tire radius r0 and the
length of the contact area LN at given wheel load, estimated values for three track points can
be given in analogy to (2.12)
b
rM L = 2
eyR r0 ezB ,
rM R = 2b eyR r0 ezB , (2.21)
LN
rM F = 2
exB r0 ezB .

The points lie left, resp. right and to the front of a point below the rim center. The unit vectors
exB and ezB point in the longitudinal and vertical direction of the vehicle. The wheel rotation
axis is given by eyR . According to (2.13) the corresponding points on the track L, R and F can
be calculated.
The vectors
rRF = r0F r0R and rRL = r0L r0R (2.22)
lie within the track plane. The unit vector calculated by
rRF rRL
en = . (2.23)
| rRF rRL |
is perpendicular to the plane defined by the points L, R, and F and gives an average track
normal over the contact area. Discontinuities which occur at step- or ramp-sized obstacles are
smoothed that way.
Of course it would be obvious to replace LN in (2.21) by the actual length L of the contact
area and the unit vector ezB by the unit vector ezR which points upwards in the wheel center

18
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

plane. The values however, can only be calculated from the current track normal. Here also an
iterative solution would be possible. Despite higher computing effort the model quality cannot
be improved by this, because approximations in the contact calculation and in the tire model
limit the exactness of the tire model.

2.2.4 Contact Point Velocity

The absolute velocity of the contact point one gets from the derivation of the position vector

v0P,0 = r0P,0 = r0M,0 + rM P,0 . (2.24)

Here r0M,0 = v0M,0 is the absolute velocity of the wheel center and rM P,0 the vector from the
wheel center M to the contact point P , expressed in the inertial frame 0. With (2.20) one gets
d
rM P,0 = (rS ezR,0 ) = rS ezR,0 rS ezR,0 . (2.25)
dt
Due to r0 = const.
rS = 4r (2.26)
follows from (2.19).
The unit vector ezR moves with the rim but does not perform rotations around the wheel rotation
axis. Its time derivative is then given by

ezR,0 = 0R,0 ezR,0 (2.27)

where 0R is the angular velocity of the wheel rim without components in the direction of the
wheel rotation axis. Now (2.25) reads as

rM P,0 = 4r ezR,0 rS 0R,0 eZR,0 (2.28)

and the contact point velocity can be written as

v0P,0 = v0M,0 + 4r ezR,0 rS 0R,0 eZR,0 . (2.29)

Because the point P lies on the track, v0P,0 must not contain a component normal to the track

eTn v0P = 0 . (2.30)

The tire deformation velocity is defined by this demand

eTn (v0M + rS 0R eZR )
4r = . (2.31)
eTn ezR
Now, the contact point velocity v0P and its components in longitudinal and lateral direction

vx = eTx v0P (2.32)

and
vy = eTy v0P (2.33)
can be calculated.

19
Vehicle Dynamics FH Regensburg, University of Applied Sciences

2.3 Wheel Load

The vertical tire force Fz can be calculated as a function of the normal tire deflection 4z =
eTn 4r and the deflection velocity 4z = eTn 4r

Fz = Fz (4z, 4z) . (2.34)

Because the tire can only deliver pressure forces to the road, the restriction Fz 0 holds.
In a first approximation Fz is separated into a static and a dynamic part

Fz = FzS + FzD . (2.35)

The static part is described as a nonlinear function of the normal tire deflection

FzS = a1 4z + a2 (4z)2 . (2.36)

The constants a1 and a2 may be calculated from the radial stiffness at nominal payload

d FzS 

cN
z = (2.37)
d 4z FzS =FzN

and the radial stiffness at double payload

d FzS 

c2N
z = . (2.38)
d 4z FzS =2FzN

The dynamic part is roughly approximated by

FzD = dR 4z , (2.39)

where dR is a constant describing the radial tire damping.

2.4 Longitudinal Force and Longitudinal Slip

To get some insight into the mechanism generating tire forces in longitudinal direction we
consider a tire on a flat test rig. The rim is rotating with the angular speed and the flat track
runs with speed vx . The distance between the rim center an the flat track is controlled to the
loaded tire radius corresponding to the wheel load Fz , Fig. 2.4.
A tread particle enters at time t = 0 the contact area. If we assume adhesion between the
particle and the track then the top of the particle runs with the track speed vx and the bottom
with the average transport velocity vt = rD . Depending on the speed difference 4v =
rD vx the tread particle is deflected in longitudinal direction

u = (rD vx ) t . (2.40)

20
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

rD

vx

rD
u
vx
L
u max

Figure 2.4: Tire on Flat Track Test Rig

The time a particle spends in the contact area can be calculated by

L
T = , (2.41)
rD ||

where L denotes the contact length, and T > 0 is assured by ||.

The maximum deflection occurs when the tread particle leaves at t = T the contact area

L
umax = (rD vx ) T = (rD vx ) . (2.42)
rD ||

The deflected tread particle applies a force to the tire. In a first approximation we get

Fxt = ctx u , (2.43)

where ctx is the stiffness of one tread particle in longitudinal direction.

On normal wheel loads more than one tread particle is in contact with the track, Fig. 2.5a. The
number p of the tread particles can be estimated by

L
p = . (2.44)
s+a
where s is the length of one particle and a denotes the distance between the particles.
Particles entering the contact area are undeformed on exit the have the maximum deflection.
According to (2.43) this results in a linear force distribution versus the contact length, Fig. 2.5b.
For p particles the resulting force in longitudinal direction is given by

1 t
Fx = p c umax . (2.45)
2 x

21
Vehicle Dynamics FH Regensburg, University of Applied Sciences

a) b) c)
L L
r0

r

s a cxt * u
cut * u max
L/2

Figure 2.5: a) Particles, b) Force Distribution, c) Tire Deformation

With (2.44) and (2.42) this results in

1 L t L
Fx = cx (rD vx ) . (2.46)
2 s+a rD ||
A first approximation of the contact length L is given by

(L/2)2 = r02 (r0 4r)2 , (2.47)

where r0 is the undeformed tire radius, and 4r denotes the tire deflection, Fig. 2.5c. With
4r  r0 one gets
L2 8 r0 4r . (2.48)
The tire deflection can be approximated by

4r = Fz /cR . (2.49)

where Fz is the wheel load, and cR denotes the radial tire stiffness. Now, (2.45) can be written
as
r0 ctx rD v x
Fx = 4 Fz . (2.50)
s + a cR rD ||
The non-dimensional relation between the sliding velocity of the tread particles in longitudinal
direction vxS = vx rD and the average transport velocity rD || forms the longitudinal slip

(vx rD )
sx = . (2.51)
rD ||
In this first approximation the longitudinal force Fx is proportional to the wheel load Fz and
the longitudinal slip sx
Fx = k Fz sx , (2.52)
where the constant k collects the tire properties r0 , s, a, ctx and cR .
The relation (2.52) holds only as long as all particles stick to the track. At average slip values
the particles at the end of the contact area start sliding, and at high slip values only the parts
at the beginning of the contact area still stick to the road, Fig. . 2.6.
The resulting nonlinear function of the longitudinal force Fx versus the longitudinal slip sx
can be defined by the parameters initial inclination (driving stiffness) dFx0 , location sM
x and
M G G
magnitude of the maximum Fx , start of full sliding sx and the sliding force Fx , Fig. 2.7.

22
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

small slip values moderate slip values large slip values

Fx = k * Fz* s x Fx = Fz * f ( s x ) Fx = FG
L L L

t t t
Fx <= FH t t Fx = FG
Fx = FH
adhesion sliding
adhesion sliding

Figure 2.6: Longitudinal Force Distribution for different Slip Values

Fx
adhesion
M sliding
Fx
G
Fx dFx0

sM
x sGx sx

Figure 2.7: Typical Longitudinal Force Characteristics

2.5 Lateral Slip, Lateral Force and Self Aligning Torque

Similar to the longitudinal slip sx , given by (2.51), the lateral slip can be defined by

vyS
sy = , (2.53)
rD ||
where the sliding velocity in lateral direction is given by

vyS = vy (2.54)

and the lateral component of the contact point velocity vy follows from (2.33).
As long as the tread particles stick to the road (small amounts of slip), an almost linear distri-
bution of the forces along the length L of the contact area appears. At moderate slip values
the particles at the end of the contact area start sliding, and at high slip values only the parts
at the beginning of the contact area stick to the road, Fig. 2.8. The nonlinear characteristics
of the lateral force versus the lateral slip can be described by the initial inclination (cornering
stiffness) dFy0 , location sM M G
y and magnitude Fy of the maximum and start of full sliding sy and
G
magnitude Fy of the sliding force.
The distribution of the lateral forces over the contact area length also defines the acting point
of the resulting lateral force. At small slip values the working point lies behind the center of

23
Vehicle Dynamics FH Regensburg, University of Applied Sciences

small slip values moderate slip values large slip values

Fy = k * Fz * s y Fy = Fz * f ( s y ) Fy = FG

adhesion
adhesion

Fy

sliding

Fy
L
L

L
Fy n

sliding
Figure 2.8: Lateral Force Distribution over Contact Area

the contact area (contact point P). With rising slip values, it moves forward, sometimes even
before the center of the contact area. At extreme slip values, when practically all particles are
sliding, the resulting force is applied at the center of the contact area.
The resulting lateral force Fy with the dynamic tire offset or pneumatic trail n as a lever gener-
ates the self aligning torque
MS = n Fy . (2.55)

The lateral force Fy as well as the dynamic tire offset are functions of the lateral slip sy . Typ-
ical plots of these quantities are shown in Fig. 2.9. Characteristic parameters for the lateral

n/L
adhesion
(n/L)0
adhesion/sliding
Fy full sliding
adhesion/
adhesion
M sliding s0y sGy sy
Fy full sliding
G
Fy dF0y MS adhesion

adhesion/sliding

full sliding

sM
y sGy sy s0y sGy sy

Figure 2.9: Typical Plot of Lateral Force, Tire Offset and Self Aligning Torque

force graph are initial inclination (cornering stiffness) dFy0 , location sM

y and magnitude of the
maximum FyM , begin of full sliding sG y , and the sliding force F G
y .
The dynamic tire offset has been normalized by the length of the contact area L. The initial
value (n/L)0 as well as the slip values s0y and sG
y characterize the graph sufficiently.

24
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

2.6 Camber Influence

At a cambered tire, Fig. 2.10, the angular velocity of the wheel has a component normal to
the road
n = sin . (2.56)

en

rim Fy = Fy (s y ): Parameter
4000
centre
plane

3000
2000
eyR n
1000
0
-1000
ex rD || -2000
-3000
v()
y()
-4000
ey -0.5 0 0.5

Figure 2.10: Cambered Tire Fy () at Fz = 3.2 kN and = 0 , 2 , 4 , 6 , 8

Now, the tread particles in the contact area possess a lateral velocity which depends on their
position and is given by
L
v () = n , = sin , L/2 L/2 . (2.57)
2 L/2
At the center of the contact area (contact point) it vanishes and at the end of the contact area
it is of the same value but opposite to the value at the beginning of the contact area.
Assuming that the tread particles stick to the track, the deflection profile is defined by

y () = v () . (2.58)

The time derivative can be transformed to a space derivative

d y () d d y ()
y () = = rD || (2.59)
d dt d
where rD || denotes the average transport velocity. Now (2.58) reads as

d y ()
rD || = sin , (2.60)
d

25
Vehicle Dynamics FH Regensburg, University of Applied Sciences

which results in the parabolic deflection profile

 2 "  2 #
1 sin L
y () = 1 . (2.61)
2 rD || 2 L/2

Similar to the lateral slip sy which is by (2.53) we now can define a camber slip

sin L
s = . (2.62)
rD || 2

The lateral deflection of the tread particles generates a lateral force

Fy = cy y , (2.63)

where cy denotes the lateral stiffness of the tread particles and

ZL/2 "  2 #
1 L 1 x 1
y = (s ) 1 d = s L (2.64)
2 2 L L/2 6
L/2

is the average value of the parabolic deflection profile.

A purely lateral tire movement without camber results in a linear deflexion profile with the
average deflexion
1
yy = sy L . (2.65)
2
A comparison of (2.64) to (2.65) shows, that with

1
sy = s (2.66)
3
the lateral camber slip s can be converted to an equivalent lateral slip sy .
In normal driving operation, 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 . (2.67)

If the global inclination dFy = Fy /sy is used instead of the initial inclination dFy0 , one gets
the camber influence on the lateral force as shown in Fig. 2.10.
The camber angle influences the distribution of pressure in the lateral direction of the contact
area, and changes the shape of the contact area from rectangular to trapezoidal. It is thus
extremely difficult if not impossible to quantify the camber influence with the aid of such simple
models. But this approach turns out to be a quit good approximation.

26
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

2.7 Bore Torque

If the angular velocity of the wheel

0W = 0R + eyR (2.68)
has a component in direction of the track normal en
n = eTn 0W 6= 0 . (2.69)
a very complicated deflection profile of the tread particles in the contact area occurs. By a
simple approach the resulting bore torque can be approximated by the parameter of the longi-
tudinal force characteristics.
Fig. 2.11 shows the contact area at zero camber, = 0 and small slip values, sx 0, sy 0.
The contact area is separated into small stripes of width dy . The longitudinal slip in a stripe at
position y is then given by
(n y)
sx (y) = . (2.70)
rD ||

dy x dy x

contact
UG area
U(y)
Q P P
L

L
y n y n -UG
contact
area

B B

Figure 2.11: Bore Torque generated by Longitudinal Forces

For small slip values the nonlinear tire force characteristics can be linearized. The longitudinal
force in the stripe can then be approximated by

d Fx  d sx
Fx (y) = y. (2.71)
d sx sx =0 d y
With (2.70) one gets 
d Fx  n
Fx (y) = y. (2.72)
d sx sx =0 rD ||

The forces Fx (y) generate a bore torque in the contact point P
B B
Z+ 2 Z+ 2 
1 1 d Fx  n
MB = y Fx (y) dy = y y dy
B B d sx sx =0 rD ||

B
2
B
2
(2.73)
 
1 2 d Fx  n 1 d Fx  B n
= B = B ,
12 d sx sx =0 rD ||
 12 d sx sx =0 rD | |


27
Vehicle Dynamics FH Regensburg, University of Applied Sciences

where
n
sB = (2.74)
||
can be considered as bore slip. Via dFx /dsx the bore torque takes into account the actual
friction and slip conditions.
The bore torque calculated by (2.73) is only a first approximation. At large bore slips the longi-
tudinal forces in the stripes are limited by the sliding values. Hence, the bore torque is limited
by
B
Z+ 2
1 1
| MB | MBmax = 2 y FxG dy = B FxG , (2.75)
B 4
0

where FxG denotes the longitudinal sliding force.

2.8 Typical Tire Characteristics

The tire model TMeasy1 which is based on this simple approach can be used for passenger
car tires as well as for truck tires. It approximates the characteristic curves Fx = Fx (sx ),
Fy = Fy () and Mz = Mz () quite well even for different wheel loads Fz , Fig. 2.12 and
Fig. 2.13.

6
40
4
20
2
F [kN]

F [kN]

0 0 10 kN
x
x

1.8 kN 20 kN
-2 3.2 kN -20 30 kN
4.6 kN 40 kN
-4 5.4 kN 50 kN
-40
-6
-40 -20 0 20 40 -40 -20 0 20 40
s [%] s [%]
x x

Figure 2.12: Longitudinal Force Characteristics: Meas., TMeasy

Within TMeasy the one-dimensional characteristics are automatically converted to a two-

dimensional combination characteristics, Fig. 2.14.

1
Hirschberg, W; Rill, G. Weinfurter, H.: User-Appropriate Tyre-Modelling for Vehicle Dynamics in Standard and
Limit Situations. Vehicle System Dynamics 2002, Vol. 38, No. 2, pp. 103-125. Lisse: Swets & Zeitlinger.

28
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

6
40
4
20
2

F [kN]
Fy [kN]

0 0

y
1.8 kN 10 kN
-2 3.2 kN -20 20 kN
4.6 kN 30 kN
-4
6.0 kN 40 kN
-40
-6
150 1500

100 1000

50 M [Nm] 500
Mz [Nm]

0 0
z

-50 1.8 kN -500

3.2 kN 18.4 kN
-100 -1000 36.8 kN
4.6 kN
6.0 kN 55.2 kN
-150 -1500
-20 -10 0 10 20 -20 -10 0 10 20
[o] [o]

Figure 2.13: Lateral Force and Self Aligning Torque: Meas., TMeasy

3 30

2 20

1 10
F [kN]

F [kN]

0 0
y

-1
-10
-2
-20
-3
-30
-4 -2 0 2 4 -20 0 20
F [kN] F [kN]
x x

|sx | = 1, 2, 4, 6, 10, 15 %; || = 1, 2, 4, 6, 10, 14

Figure 2.14: Two-dimensional Tire Characteristics at Fz = 3.2 kN / Fz = 35 kN

29
3 Longitudinal Dynamics

3.1 Dynamic Wheel Loads

3.1.1 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 into one normal and
one longitudinal force.

h mg

Fx1 Fx2
Fz1 a1 a2 Fz2

Figure 3.1: Simple Vehicle Model

If aerodynamic forces (drag, positive and negative lift) are neglected at first, then the equations
of motions in the x-, z -plane read as

m v = Fx1 + Fx2 , (3.1)

0 = Fz1 + Fz2 m g (3.2)

and
0 = Fz1 a1 Fz2 a2 + (Fx1 + Fx2 ) h , (3.3)
where v indicates the vehicles acceleration, m is the mass of the vehicle, a1 +a2 is the wheel
base, and h is the height of the center of gravity.

30
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

This are only three equations for the four unknown forces Fx1 , Fx2 , Fz1 , Fz2 . But, if we insert
(3.1) in (3.3) we can eliminate two unknowns by one stroke

0 = Fz1 a1 Fz2 a2 + m v h . (3.4)

The equations (3.2) and (3.4) can now be resolved for the axle loads
 
a2 h v
Fz1 = m g (3.5)
a1 + a2 a1 + a2 g

and  
a1 h v
Fz2 = m g + . (3.6)
a1 + a2 a1 + a2 g
The weight of the vehicle mg is distributed among the axles according to position of the cen-
ter of gravity. When accelerating v > 0, the front axle is relieved, as is the rear axle when
decelerating v < 0.

3.1.2 Influence of Grade

For a vehicle on a grade, Fig.3.2, the equations of motions (3.1) to (3.3) can easily be extended
to
m v = Fx1 + Fx2 m g sin ,
0 = Fz1 + Fz2 m g cos , (3.7)
0 = Fz1 a1 Fz2 a2 + (Fx1 + Fx2 ) h ,
where denotes the grade angle.

v z

Fx1
mg
h
Fz1 a1
Fx2
a2
Fz2

Figure 3.2: Vehicle on Grade

31
Vehicle Dynamics FH Regensburg, University of Applied Sciences

Now, the dynamic axle loads are given by

   
a2 h v
Fz1 = m g cos + tan (3.8)
a1 + a2 a1 + a2 g
and    
a1 h v
Fz2 = m g cos + + tan . (3.9)
a1 + a2 a1 + a2 g
A grade angle 6= 0 reduces the static load, mg mg cos , and generates an additional
acceleration term, vg vg + tan .

3.1.3 Aerodynamic Forces

The shape of most vehicles or specific wings mounted at the vehicle produce aerodynamic
forces and torques. The effect of this 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. 3.3.

FD1 FD2

FAR

h
mg
Fx1 Fx2
a1 a2
Fz1 Fz2

Figure 3.3: Vehicle with Aerodynamic Forces

If we assume a positive driving speed, v >, then the equations of motion read as
m v = Fx1 + Fx2 FAR ,
0 = Fz1 FD1 + Fz2 FD2 m g , (3.10)
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
   
a2 h v FAR
Fz1 = FD1 + m g + (3.11)
a1 + a2 a1 + a2 g mg
and    
a1 h v FAR
Fz2 = FD2 + m g + + . (3.12)
a1 + a2 a1 + a2 g mg
The down forces FD1 , FD2 increase the axle loads, and the air resistance FAR generates an
additional acceleration term, vg vg + FmAR
g
.

32
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

3.2 Maximum Acceleration

3.2.1 Tilting Limits

Ordinary automotive vehicles can only deliver pressure forces to the road. If we apply the
demands Fz1 0 and Fz2 0 to (3.8) and (3.9) we get
a1 v a2
+ tan . (3.13)
h g h
This tilting condition limits the acceleration v and by the grade angle the climbing capacity
of the vehicle.
The presence of aerodynamic forces complicates the tilting condition. Aerodynamic forces
become important only at high speeds. Here the vehicle acceleration normally is limited by the
engine power.

3.2.2 Friction Limits

The maximum acceleration is also limited by the friction conditions

|Fx1 | Fz1 and |Fx2 | Fz2 (3.14)

where the same friction coefficient has been assumed at front and rear axle.
In the limit case
Fx1 = Fz1 and Fx2 = Fz2 (3.15)
the first equation in (3.7) can be written as

m vmax = (Fz1 + Fz2 ) m g sin . (3.16)

Using (3.8) and (3.9) one gets

 
v
= cos sin . (3.17)
g max
That means climbing (v > 0, > 0) or downhill stopping (v < 0, < 0) requires at least a
friction coefficient tan .
According to the vehicle dimensions and the friction values the maximal acceleration or decel-
eration is restricted either by (3.13) or by (3.17).
If we take aerodynamic forces into account the maximum acceleration on a horizontal road is
limited by
   
FD1 FD2 FAR v FD1 FD2 FAR
1 + + 1 + + . (3.18)
mg mg mg g mg mg mg
In particular the aerodynamic forces enhance the braking performance of the vehicle.

33
Vehicle Dynamics FH Regensburg, University of Applied Sciences

3.3 Driving and Braking

3.3.1 Single Axle Drive

With the rear axle driven in limit situations Fx1 = 0 and Fx2 = Fz2 holds. Then, using (3.6) the
linear momentum (3.1) results in
 
a1 h vR WD
m vR WD = m g + , (3.19)
a1 + a2 a1 + a2 g
where the subscript R WD indicates the rear wheel drive. Hence, the maximum acceleration for
a rear wheel driven vehicle is given by
vR WD a1
= . (3.20)
g h a1 + a2
1
a1 + a2

By setting Fx1 = Fz1 and Fx2 = 0 the maximum acceleration for a front wheel driven vehicle
can be calculated in a similar way. One gets
vF WD a2
= , (3.21)
g h a1 + a2
1+
a1 + a2
where the subscript F WD denotes front wheel drive. Depending on the parameter , a1 , a2 and
h the accelerations may be limited by the tilting condition vg ah2 .
The maximum accelerations of a single axle driven vehicle are plotted in Fig. 3.4.

range of load distribution

1
RWD

FWD

FWD
0.8
v/g

.
0.6

0.4

RWD
0.2

0
0 0.2 0.4 0.6 0.8 1
a2 / (a1+a2)

Figure 3.4: Single Axle Driven Passenger Car: = 1, h = 0.55 m, a1 +a2 = 2.5 m

For rear wheel driven passenger cars the parameter 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

34
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

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 faster than front
wheel driven vehicles.

3.3.2 Braking at Single Axle

If only the front axle is braked then in the limit case Fx1 = Fz1 and Fx2 = 0 holds. With (3.5)
one gets from (3.1)
 
a2 h vF WB
m vF WB = m g (3.22)
a1 + a2 a1 + a2 g
where the subscript F WB indicates front wheel braking. The maximum deceleration is then
given by
vF WB a2
= . (3.23)
g h a1 + a2
1
a1 + a2
If only the rear axle is braked (Fx1 = 0, Fx2 = Fz2 ) one gets the maximal deceleration
vR WB a1
= , (3.24)
g h a1 + a2
1+
a1 + a2
where the subscript R WB indicates a braked rear axle. Depending on the parameter , a1 , a2
and h the decelerations may be limited by the tilting condition vg ah1 .
The maximum decelerations of a single axle braked vehicle are plotted in Fig. 3.5.

0
range of
load
distribution
-0.2
v/g

FWB
.
-0.4

-0.6

-0.8
RWB
-1
0 0.2 0.4 0.6 0.8 1
a2 / (a1+a2)

Figure 3.5: Single Axle Braked Passenger Car: = 1, h = 0.55 m, a1 +a2 = 2.5 m

For passenger cars the load distribution parameter a2 /(a1 +a2 ) usually covers the range from
0.4 to 0.6. If only the front axle is braked then decelerations from v/g = 0.51 to v/g = 0.77

35
Vehicle Dynamics FH Regensburg, University of Applied Sciences

can be achieved. This is pretty much compared to the deceleration range of a braked rear axle
which is in the range from v/g = 0.49 to v/g = 0.33.
That is why the braking system at the front axle has a redundant design.

3.3.3 Optimal Distribution of Drive and Brake Forces

The sum of the longitudinal forces accelerates or decelerates the vehicle. In dimensionless
style (3.1) reads
v Fx1 Fx2
= + . (3.25)
g mg mg
A certain acceleration or deceleration can only be achieved by different combinations of the
longitudinal forces Fx1 and Fx2 . According to (3.15) the longitudinal forces are limited by wheel
load and friction.
The optimal combination of Fx1 and Fx2 is achieved, when front and rear axle have the same
skid resistance.
Fx1 = Fz1 and Fx2 = Fz2 . (3.26)
With (3.5) and (3.6) one gets
 
Fx1 a2 v h
= (3.27)
mg h g a1 + a2

and  
Fx2 a1 v h
= + . (3.28)
mg h g a1 + a2
With (3.27) and (3.28) one gets from (3.25)

v
= , (3.29)
g
where it has been assumed that Fx1 and Fx2 have the same sign.
With (3.29 inserted in (3.27) and (3.28) one gets
 
Fx1 v a2 v h
= (3.30)
mg g h g a1 + a2

and  
Fx2 v a1 v h
= + . (3.31)
mg g h g a1 + a2
remain.
Depending on the desired acceleration v > 0 or deceleration v < 0 the longitudinal forces that
grant the same skid resistance at both axles can now be calculated.
Fig.3.6 shows the curve of optimal drive and brake forces for typical passenger car values. At

36
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

B2/mg
braking
Fx1/mg -a1/h

dFx2
0 -1 -2 B1/mg
dFx1 0

a =1.15
1
a =1.35
driving

2
1 h=0.55
=1.20

tilting limits

2
a2/h

Fx2/mg

Figure 3.6: Optimal Distribution of Drive and Brake Forces

the tilting limits v/g = a1 /h and v/g = +a2 /h no longitudinal forces can be delivered at the
lifting axle.
The initial gradient only depends on the steady state distribution of wheel loads. From (3.30)
and (3.31) it follows
Fx1
d  
mg a2 v h
= 2 (3.32)
v h g a1 + a2
d
g
and
Fx2
d  
mg a1 v h
= +2 . (3.33)
v h g a1 + a2
d
g
For v/g = 0 the initial gradient remains as

d Fx2  a1
= . (3.34)
d Fx1 0 a2

37
Vehicle Dynamics FH Regensburg, University of Applied Sciences

3.3.4 Different Distributions of Brake Forces

In practice it is tried to approximate the optimal distribution of brake forces by constant distri-
bution, limitation or reduction of brake forces as good as possible. Fig. 3.7.

Fx1/mg Fx1/mg Fx1/mg

constant

Fx2/mg

Fx2/mg
Fx2/mg

limitation reduction
distribution

Figure 3.7: Different Distributions of Brake Forces

When braking, the vehicles stability is dependent on the potential of lateral force (cornering
stiffness) at the rear axle. In practice, a greater skid (locking) resistance is thus realized at the
rear axle than at the front axle. Because of this, the brake force balances in the physically rel-
evant area are all below the optimal curve. This restricts the achievable deceleration, specially
at low friction values.
Because the optimal curve is dependent on the vehicles center of gravity additional safeties
have to be installed when designing real distributions of brake forces.
Often the distribution of brake forces is fitted to the axle loads. There the influence of the height
of the center of gravity, which may also vary much on trucks, remains unrespected and has to
be compensated by a safety distance from the optimal curve.
Only the control of brake pressure in anti-lock-systems provides an optimal distribution of brake
forces independent from loading conditions.

3.3.5 Anti-Lock-Systems

Lateral forces can only be scarcely transmitted, if high values of longitudinal slip occur when
decelerating a vehicle. Stability and/or steerability is then no longer given.
By controlling the brake torque, respectively brake pressure, the longitudinal slip can be re-
stricted to values that allow considerable lateral forces.
The angular wheel acceleration is used here as control variable. Angular wheel accelerations
are derived from the measured angular wheel speeds by differentiation. With a longitudinal slip
of sL = 0 the rolling condition is fulfilled. Then

rD = x (3.35)

holds, where rD labels the dynamic tyre radius and x is the vehicles acceleration. According
to (3.17), the maximum acceleration/deceleration of a vehicle is dependent on the friction
coefficient, |x| = g . With a known friction coefficient a simple control law can be realized
for every wheel
1
|| |x| . (3.36)
rD

38
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

Because until today no reliable possibility to determine the local friction coefficient between
tyre and road has been found, useful information can only be gained from (3.36) 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 then used for the calculation of slip at all wheels. This method is too imprecise
at low speeds. Below a limit velocity no control occurs therefore. Problems also occur when for
example all wheels lock simultaneously which may happen on icy roads.
The control of the brake torque is done via the brake pressure which can be increased, held or
decreased by a three-way valve. To prevent vibrations, the decrement is usually made slower
than the increment.
To prevent a strong yaw reaction, the select low principle is often used with -split braking
at the rear axle. The break pressure at both wheels is controlled the wheel running on lower
friction. Thus the brake forces at the rear axle cause no yaw torque. The maximally achievable
deceleration however is reduced by this.

3.4 Drive and Brake Pitch

3.4.1 Vehicle Model

zA

xA
MA1 FF1
A
z1
MA1 MB1
hR FF2
MA2
R1
MB1 z2

Fz1 Fx1 MA2 MB2

a1 R R2
MB2

a2
Fz2 Fx2

Figure 3.8: Plane Vehicle Model

The vehicle model drawn in Fig. 3.8 consists of five rigid bodies. The body has three degrees
of freedom: Longitudinal motion xA , vertical motion zA and pitch A . The coordinates z1 and z2

39
Vehicle Dynamics FH Regensburg, University of Applied Sciences

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 ) ;
(3.37)
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 labelled FF 1 and FF 2 . At the wheels
drive torques MA1 , MA2 and brake torques MB1 , MB2 , longitudinal forces Fx1 , Fx2 and the
wheel loads Fz1 , Fz2 apply. The brake torques are supported directly by the wheel bodies, the
drive torques are transmitted by the drive shafts to the vehicle body. The forces and torques
that apply to the single bodies are listed in the last column of the tables 3.1 and 3.2.
The velocity of the vehicle body and its angular velocity is given by

xA 0 0
v0A,0 = 0 + 0 ; 0A,0 = A . (3.38)
0 zA 0
At small rotational motions of the body one gets for the speed of the wheel bodies and wheels
x1

xA 0 hR A z1
z1
v0RK1 ,0 = v0R1 ,0 = 0 + 0 + 0 + 0 ; (3.39)
0 zA a1 A z1
x2

xA 0 hR A z2
z2
v0RK2 ,0 = v0R2 ,0 = 0 + 0 + 0 + 0 . (3.40)
0 zA +a2 A z2
The angular velocities of the wheel bodies and wheels are given by

0 0 0 0 0
0RK1 ,0 = A + 1 and 0R1 ,0 = A + 1 + R1
0 0 0 0 0
(3.41)
as well as

0 0 0 0 0
0RK2 ,0 = A + 2 and 0R2 ,0 = A + 2 + R2
0 0 0 0 0
(3.42)
Introducing a vector of generalized velocities
 T
z = xA zA A 1 R1 2 R2 (3.43)
the velocities and angular velocities (3.38), (3.39), (3.40), (3.41), (3.42) can be written as
7 7
X v0i X 0i
v0i = zj and 0i = zj (3.44)
j=1
zj j=1
zj

40
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

3.4.2 Equations of Motion

The partial velocities v

zj
0i
and partial angular velocities 0i
zj
for the five bodies i = 1(1)5 and
for the seven generalized speeds j = 1(1)7 are arranged in the tables 3.1 and 3.2. With the

partial velocities v0i /zj applied forces

bodies xA zA A z1 R1 z2 R2 Fie
1 0 0 0 0 0 0 0
chassis 0 0 0 0 0 0 0 0
mA
0 1 0 0 0 0 0 FF 1 +FF 2 mA g
x1
wheel body 1 0 hR z1 0 0 0 0
front 0 0 0 0 0 0 0 0
mRK1 0 1 a1 1 0 0 0 FF 1 mRK1 g
x1
wheel 1 0 hR z1 0 0 0 Fx1
front 0 0 0 0 0 0 0 0
mR1 0 1 a1 1 0 0 0 Fz1 mR1 g
x2
wheel body 1 0 hR 0 0 z2 0 0
rear 0 0 0 0 0 0 0 0
mRK2 0 1 a2 0 0 1 0 FF 2 mRK2 g
x2
wheel 1 0 hR 0 0 z2 0 Fx2
rear 0 0 0 0 0 0 0 0
mR2 0 1 a2 0 0 1 0 Fz2 mR2 g

Table 3.1: Partial Velocities and Applied Forces

partial angular velocities 0i /zj applied torques

bodies xA zA A z1 R1 z2 R2 Mie
0 0 0 0 0 0 0 0
chassis 0 0 1 0 0 0 0 MA1MA2 a1 FF 1 +a2 FF 2
A
0 0 0 0 0 0 0 0
wheel body 0 0 0 0 0 0 0 0
1
front 0 0 1 z1
0 0 0 MB1
RK1 0 0 0 0 0 0 0 0
wheel 0 0 0 0 0 0 0 0
1
front 0 0 1 z1
1 0 0 MA1 MB1 R Fx1
R1 0 0 0 0 0 0 0 0
wheel body 0 0 0 0 0 0 0 0
2
rear 0 0 1 0 0 z2
0 MB2
RK2 0 0 0 0 0 0 0 0
wheel 0 0 0 0 0 0 0 0
2
rear 0 0 1 0 0 z2
1 MA2 MB2 R Fx2
R2 0 0 0 0 0 0 0 0

Table 3.2: Partial Angular Velocities and Applied Torques

41
Vehicle Dynamics FH Regensburg, University of Applied Sciences

aid of the partial velocities and partial angular velocities the elements of the mass matrix M
and the components of the vector of generalized forces and torques Q can be calculated.
5  T 5  T
X v0k v0k X 0k 0k
M (i, j) = mk + k ; i, j = 1(1)7 ; (3.45)
k=1
zi zj k=1
zi zj
5  T 5  T
X v0k X 0k
Q(i) = Fke + Mke ; i = 1(1)7 . (3.46)
k=1
zi k=1
zi

The equations of motion for the plane vehicle model are then given by

M z = Q . (3.47)

3.4.3 Equilibrium

With the abbreviations

m1 = mRK1 + mR1 ; m2 = mRK2 + mR2 ; mG = mA + m1 + m2 (3.48)

and
h = hR + R (3.49)
The components of the vector of generalized forces and torques read as

Q(1) = Fx1 + Fx2 ;

Q(2) = Fz1 + Fz2 mG g ; (3.50)

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

x1 1
Q(4) = Fz1 FF 1 + z1
Fx1 m1 g + z1
(MA1 R Fx1 ) ;
(3.51)
Q(5) = MA1 MB1 R Fx1 ;
x2 2
Q(6) = Fz2 FF 2 + z2
Fx2 m2 g + z2
(MA2 R Fx2 ) ;
(3.52)
Q(7) = MA2 MB2 R Fx2 .

Without drive and brake forces

MA1 = 0 ; MA2 = 0 ; MB1 = 0 ; MB2 = 0 (3.53)

from (3.50), (3.51) and (3.52) one gets the steady state longitudinal forces, the spring preloads
and the wheel loads
0 0
Fx1 = 0; Fx2 = 0;
b a
FF0 1 = a+b
mA g ; FF0 2 = a+b
mA g ; (3.54)
0 b 0 a
Fz1 = m1 g + a+b
mA g ; Fz2 = m2 g + a+b
mA g .

42
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

3.4.4 Driving and Braking

Assuming that on accelerating or decelerating the vehicle xA 6= 0 the wheels neither slip nor
lock,
x1
R R1 = xA hR A + z1
z1 ;
x2
(3.55)
R R2 = xA hR A + z2
z2 .
holds. 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. (3.56)

and (3.55) simplifies to

R R1 = xA ; R R2 = xA . (3.57)
With(3.56), (3.57) and (3.49) the equation of motion (3.47) results in

a a
mG xA = Fx1 + Fx2 ;
a a
0 = Fz1 + Fz2 ;
xA xA a a a a
hR (m1 +m2 ) xA + R1 R
+ R2 R
= a Fz1 + b Fz2 (hR + R)(Fx1 + Fx2 );
(3.58)
1 1
x1
z1
m1 xA + z1
R1 xRA = a
Fz1 FFa 1 + x1
z1
a
Fx1 + z1
(MA1 a
R Fx1 ) ;
(3.59)
xA a
R1 R
= MA1 MB1 R Fx1 ;

x2 2 xA a x2 2
z2
m2 xA + z2
R2 R
= Fz2 FFa 2 + z2
a
Fx2 + z2
(MA2 a
R Fx2 );
(3.60)
xA a
R2 R
= MA2 MB2 R Fx2 ;
where the steady state spring forces, longitudinal forces and wheel loads have been separated
into initial and acceleration-dependent terms
st 0 a
Fxi = Fxi + Fxi ; Fzist = Fzi0 + Fzia ; FFsti = FF0 i + FFa i ; i = 1, 2 . (3.61)

a a
With given torques of drive and brake the vehicle acceleration xA , the wheel forces Fx1 , Fx2 ,
a a a a
Fz1 , Fz2 and the spring forces FF 1 , FF 2 can be calculated from (3.58), (3.59) and (3.60)
Via the spring characteristics which have been assumed as linear the acceleration-dependent
forces also cause a vertical displacement and pitch motion of the body

FFa 1 = cA1 z1a ,

FFa 2 = cA2 z2a ,
a
(3.62)
Fz1 = cR1 (zAa a Aa + z1a ) ,
a
Fz2 = cR2 (zAa + b Aa + z2a ) .
besides the vertical motions of the wheels.

43
Vehicle Dynamics FH Regensburg, University of Applied Sciences

a
Especially the pitch of the vehicle A 6= 0, caused by drive or brake is, if too distinct, felt as
annoying.
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 jounce and rebound.

3.4.5 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. 3.9.

pitch pole

x-, z- motion of the contact points

during compression and rebound

Figure 3.9: Brake Pitch Pole

Increasing the pitch pole height above the track level means a decrease in the brake pitch
angle.

44
4 Lateral Dynamics

4.1 Kinematic Approach

4.1.1 Kinematic Tire Model

When a vehicle drives through the curve at low lateral acceleration, low lateral forces are
needed for course holding. At the wheels then hardly lateral slip occurs. In the ideal case, with
vanishing lateral slip, the wheels only move in circumferential direction. The speed component
of the contact point in the tires lateral direction then vanishes
vy = eTy v0P = 0 . (4.1)
This kinematic constraint equation can be used for course calculation of slowly moving vehi-
cles.

4.1.2 Ackermann Geometry

Within the validity limits of the kinematic tire model the necessary steering angle of the front
wheels can be constructed via given momentary turning center M , Fig. 4.1.
At slowly moving vehicles the lay out of the steering linkage is usually done according to the
Ackermann geometry. With given steering angle of a wheel, e.g. 2 , the wheel base a fixes the
momentarily driven curve radius
a a
tan 2 = or R= . (4.2)
R tan 2
For the second wheel however
a
tan 1 = (4.3)
R+b
holds, with b describing the track width. Combining (4.2) and (4.3) we get
a tan 2
tan 1 = a or tan 1 = . (4.4)
+b b
tan 2 1 + tan 2
a
The deviations 41 = 1a 1A of the actual steering angle 1a from the Ackermann steering
angle 1A , which follows from (4.4), are used to judge a steering system.
At a rotation around the momentary pole M the direction of the velocity is fixed for every point
of the vehicle. The angle between the velocity vector v and the vehicles longitudinal axis is
called side slip angle.

45
Vehicle Dynamics FH Regensburg, University of Applied Sciences

v
a

1
2
M

b R

Figure 4.1: Ackermann Steering Geometry at a two-axled Vehicle

4.1.3 Space Requirement

The Ackermann approach can also be used to calculate the space requirement of a vehicle
during cornering, Fig. 4.2.
If the front wheels of a two-axled vehicle are steered according to the Ackermann geometry
the outer point of the vehicle front runs on the maximum radius Rmax and a point on the inner
side of the vehicle at the location of the rear axle runs on the minimum radius Rmin . We get
2
Rmax = (Rmin + b)2 + (a + f )2 , (4.5)

where a, b are the wheel base and the width of the vehicle, and f specifies the distance of the
vehicle front to the front axle. Hence, the space requirement
q
4R = Rmax Rmin = (Rmin + b)2 + (a + f )2 Rmin , (4.6)

can be calculated as a function of the cornering radius Rmin .

The space requirement 4R of a typical passenger car and a bus is plotted in Fig. 4.3 versus
the minimum cornering radius.
In narrow curves Rmin = 5.0 m a bus requires a space of 2.5 the width, whereas a passenger
car needs only 1.5 the width.

46
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

R
f ma
x
a

Rmin
M

Figure 4.2: Space Requirement

7
bus: a=6.25 m, b=2.50 m, f=2.25 m
6 car: a=2.50 m, b=1.60 m, f=1.00 m

4
R [m]

0
0 10 20 30 40 50
R min [m]

Figure 4.3: Space Requirement of typical Passenger Car and Bus

47
Vehicle Dynamics FH Regensburg, University of Applied Sciences

4.1.4 Vehicle Model with Trailer

4.1.4.1 Position

Fig. 4.4 shows a simple lateral dynamics model for a two-axled vehicle with a single-axled
trailer. Vehicle and trailer move on a horizontal track. The position and the orientation of the

1
x
1
y
a
A1
y2
b
x2
K A2
c

y0

3
x
y3

A3 x0

Figure 4.4: Kinematic Model with Trailer

vehicle relative to the track fixed frame x0 , y0 , z0 is defined by the position vector to the rear
axle center
xF
r0A2 ,0 = yF (4.7)

R
and the rotation matrix
cos sin 0
A02 = sin cos 0 . (4.8)
0 0 1
Here, the tire radius R is considered to be constant, and xF , yF as well as are generalized
coordinates.

48
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

The position vector

a
r0A1 ,0 = r0A2 ,0 + A02 rA2 A1 ,2 mit rA2 A1 ,2 = 0 (4.9)
0

and the rotation matrix

cos sin 0
A01 = A02 A21 mit A21 = sin cos 0 (4.10)
0 0 1

describe the position and the orientation of the front axle, where a = const labels the wheel
base and the steering angle.
The position vector

r0A3 ,0 = r0A2 ,0 + A02 rA2 K,2 + A23 rKA3 ,3 (4.11)
with
b c
rA2 K,2 = 0 and rKA3 ,2 = 0 (4.12)
0 0
and the rotation matrix

cos sin 0
A03 = A02 A23 mit A23 = sin cos 0 (4.13)
0 0 1

define the position and the orientation of the trailer axis, with labelling the bend angle be-
tween vehicle and trailer and b, c marking the distances from the rear axle A2 to the coupling
point K and from the coupling point K to the trailer axis A3 .

4.1.4.2 Vehicle

According to the kinematic tire model, cf. section 4.1.1, the velocity at the rear axle can only
have a component in the vehicles longitudinal direction

vx2
v0A2 ,2 = 0 . (4.14)
0

The time derivative of (4.7) results in

xF
v0A2 ,0 = r0A2 ,0 = yF . (4.15)
0

49
Vehicle Dynamics FH Regensburg, University of Applied Sciences

With the transformation of (4.14) into the system 0

vx2 cos vx2
v0A2 ,0 = A02 v0A2 ,2 = A02 0 = sin vx2 (4.16)
0 0

one gets by equalizing with (4.15) two first order differential equations for the position coordi-
nates xF and yF
xF = cos vx2 ,
(4.17)
yF = sin vx2 .
The velocity at the front axis follows from (4.9)

v0A1 ,0 = r0A1 ,0 = r0A2 ,0 + 02,0 A02 rA2 A1 ,2 . (4.18)

Transformed into the vehicle fixed system x2 , y2 , z2

vx2 0 a vx2
v0A1 ,2 = 0 + 0 0 = a . (4.19)
0 0 0
| {z } | {z } | {z }
v0A2 ,2 02,2 rA2 A1 ,2

remains. The unit vectors

cos sin
ex1,2 = sin and ey1,2 = cos (4.20)
0 0

define the longitudinal and lateral direction at the front axle.

According to (4.1) the velocity component lateral to the wheel must vanish,

eTy1,2 v0A1 ,2 = sin vx2 + cos a = 0 . (4.21)

In longitudinal direction then

eTx1,2 v0A1 ,2 = cos vx2 + sin a = vx1 (4.22)

remains.
From (4.21) a first order differential equation follows for the yaw angle

vx2 (4.23)
= tan .
a

50
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

4.1.4.3 Entering a Curve

In analogy to (4.2) the steering angle can be related to the current track radius R or with
k = 1/R to the current track curvature
a
tan = = ak . (4.24)
R
The differential equation for the yaw angle then reads as

= vx2 k . (4.25)

With the curvature gradient

t
k = k(t) = kC (4.26)
T
The entering of a curve is described as a continuous transition from a line with the curvature
k = 0 into a circle with the curvature k = kC .
The yaw angle of the vehicle can now be calculated by simple integration

vx2 kC t2
(t) = , (4.27)
T 2
where at time t = 0 a vanishing yaw angle, (t = 0) = 0, has been assumed.
The vehicles position then follows with (4.27) from the differential equations (4.17)

Zt=T Zt=T
vx2 kC t2 vx2 kC t2
   
xF = vx2 cos dt , yF = vx2 sin dt . (4.28)
T 2 T 2
t=0 t=0

At constant vehicle speed vx2 = const. (4.28) is the parameterized form of a clothoide.
From (4.24) the necessary steering angle can be calculated, too. If only small steering an-
gles are necessary for driving through the curve, the tan-function can be approximated by its
argument, and
t
= (t) a k = a kC (4.29)
T
holds, i.e. the driving through a clothoide is manageable by continuous steer motion.

4.1.4.4 Trailer

The velocity of the trailer axis can be received by differentiation of the position vector (4.11)

v0A3 ,0 = r0A3 ,0 = r0A2 ,0 + 02,0 A02 rA2 A3 ,2 + A02 rA2 A3 ,2 . (4.30)

With
b c cos
rA2 A3 ,2 = rA2 K,2 + A23 rKA3 ,3 = c sin (4.31)
0

51
Vehicle Dynamics FH Regensburg, University of Applied Sciences

and
0 c cos c sin
rA2 A3 ,2 = 0 c sin = c cos (4.32)
0 0
| {z } | {z }
23,2 A23 rKA3 ,3
it remains, if (4.30) is transformed into the vehicle fixed frame x2 , y2 , z2

vx2 0 b c cos c sin vx2 + c sin (+ )
v0A3 ,2 = 0 + 0 c sin + c cos = b c cos (+ ) .
0 0 0 0
| {z } | {z } | {z } | {z }
v0A2 ,2 02,2 rA2 A3 ,2 rA2 A3 ,2
(4.33)
The longitudinal and lateral direction at the trailer axis are defined by the unit vectors

cos sin
ex3,2 = sin and ey3,2 = cos . (4.34)
0 0

At the trailer axis the lateral velocity must also vanish

eTy3,2 v0A3 ,2 = sin vx2 + c sin (+ ) + cos b c cos (+ ) = 0 . (4.35)

In longitudinal direction

eTx3,2 v0A3 ,2 = cos vx2 + c sin (+ ) + sin b c cos (+ ) = vx3 (4.36)

remains.
When (4.23) is inserted into (4.35), one gets a differential equation of first order for the bend
angle    
vx2 a b
= sin + cos + 1 tan . (4.37)
a c c
The differential equations (4.17) and (4.23) describe position and orientation within the x0 , y0
plane. The position of the trailer relative to the vehicle follows from (4.37).

4.1.4.5 Course Calculations

For a given set of vehicle parameters a, b, c, and predefined time functions of the vehicle
speed, vx2 = vx2 (t) and the steering angle, = (t) the course of vehicle and trailer can be
calculated by numerical integration of the differential equations (4.17), (4.23) and (4.37).
If the steering angle is slowly increased at constant driving speed, then the vehicle drives figure
which is similar to a clothoide, Fig. 4.5.

52
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

20 front axle
rear axle
trailer axle
[m]

10

0
-30 -20 -10 0 10 20 30 40 50 60
[m]

30
front axle steer angle
20
[Grad]

10

0
0 5 10 15 20 25 30
[s]

Figure 4.5: Entering a Curve

4.2 Steady State Cornering

4.2.1 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 deflection and the body roll have to
be respected., Fig. 4.6.
The balance of torques at the already inclined vehicle delivers for small angles 1  1, 2  1
s
(FzL FzR ) = m ay (h1 + h2 ) + m g [(h1 + h2 )1 + h2 2 ] , (4.38)
2
where ay indicates the lateral acceleration and m is the sprung mass.
On a left-hand tilt, the right tire raises
K
FzR = 0 (4.39)
and the left tire carries all the vehicle weight
K
FzL = mg . (4.40)

Using (4.39) and (4.40) one gets from (4.38)

s
aK
y 2 h2
= 1K K . (4.41)
g h1 + h2 h1 + h2 2

53
Vehicle Dynamics FH Regensburg, University of Applied Sciences

2 1

m ay

h2
mg

h1

F yL F yR
FzR
FzL s/2
s/2

Figure 4.6: Overturning Hazard on Trucks

The vehicle turns over, when the lateral acceleration ay rises above the limit aK
y

Roll of axle and body reduce the overturning limit. The angles 1K and 2K can be calculated
from the tire stiffness cR and the bodys roll stiffness.
On a straight-ahead drive, the vehicle weight is equally distributed to both tires

stat stat 1
FzR = FzL = mg . (4.42)
2
With
K stat
FzL = FzL + 4Fz (4.43)
and the relations (4.40), (4.42) one gets for the increase of the wheel load at the overturning
limit
1
4Fz = mg . (4.44)
2
The resulting tire deflection then follows from

4Fz = cR 4r , (4.45)

where cR is the radial tire stiffness.

54
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

Because the right tire simultaneously rebounds for the same amount, for the roll angle of the
axle
2 4r mg
2 4r = s 1K or 1K = = . (4.46)
s s cR
holds.
In analogy to (4.38) the balance of torques at the body delivers

cW 2 = m ay h2 + m g h2 (1 + 2 ) , (4.47)

where cW names the roll stiffness of the body suspension.

Accordingly, at the overturning limit ay = aK
y

aK
y mgh2 mgh2
2K = + K (4.48)
g cW mgh2 cW mgh2 1

holds. Not allowing the vehicle to overturn already at aK

y = 0 demands a minimum of roll
min
stiffness cW > cW = mgh2 .
With (4.46) and (4.48) the overturning condition (4.41) reads as

aK
y s 1 aK
y 1 1 1
(h1 + h2 ) = (h1 + h2 ) h2
h2 , (4.49)
g 2 cR g cW 1 cW 1 cR

where, for abbreviation purposes, the dimensionless stiffnesses

cR cW
cR = m g and cW = (4.50)
m g h2
s
have been used.
Resolved for the normalized lateral acceleration
s
aK
y 2 1
= (4.51)
g h2 cR
h1 + h2 +
cW 1

remains.
At heavy trucks, a twin tire axle can be loaded with m = 13 000 kg . The radial stiffness of one
tire is cR = 800 000 N/m and the track with can be set to s = 2 m. The values h1 = 0.8 m and
h2 = 1.0 m hold at maximal load. This values deliver the results shown in Fig. 4.7 Even at a
rigid body suspension cW the vehicle turns over at a lateral acceleration of ay 0.5 g .
The roll angle of the vehicle then solely results from the tire deflection.
At a normalized roll stiffness of cW = 5 the overturning limit lies at ay 0.45 g and so reaches
already 90% of the maximum. The vehicle will then turn over at a roll angle of 10 .

55
Vehicle Dynamics FH Regensburg, University of Applied Sciences

overturning limit a y /g roll angle =K +K

1 2
0.6 20

0.5
15
0.4

0.3 10

0.2
5
0.1

0 0
0 10 20 0 10 20
*
normalized roll stiffness c W *
normalized roll stiffness c W

Figure 4.7: Tilting Limit for a Truck at Steady State Cornering

4.2.2 Roll Support and Camber Compensation

When a vehicle drives through a curve with the lateral acceleration ay , centrifugal forces are
delivered to the single masses. At the even roll model in Fig. 4.8 these are the forces mA ay
and mR ay , where mA names the body mass and mR the wheel mass.
Through the centrifugal force mA ay applied to the body at the center of gravity, a roll torque is
generated, that rolls the body with the angle A and leads to a counterwise deflection of the
tires z1 = z2 .

b/2 b/2

zA A

mA a y SA yA

FF2 FF1
h0
z2 2 z1 1
mR a y S2 mR a y S1
r0 y2 y1
Q2 Q1
F y2 F y1
Fy2 F z1

Figure 4.8: Plane Roll Model

56
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

At steady state cornering, the vehicle is balanced. With the principle of virtual work
W = 0 (4.52)
the equilibrium position can be calculated.
At the plane vehicle model in Fig. 4.8 the suspension forces FF 1 , FF 2 and tire forces Fy1 , Fz1 ,
Fy2 , Fz2 , are approximated by linear spring elements with the constants cA and cQ , cR . The
work W of these forces can be calculated directly or using W = V via the potential V . At
small deflections 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
(4.53)
12 cS (z1 z2 )2
12 cQ (yA + h0 A + y1 + r0 1 )2 12 cQ (yA + h0 A + y2 + r0 2 )2

2
2
12 cR zA + 2b A + z1 12 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
y y
y1 = z1 , 1 = 1 and y2 = z2 , 2 = 2 (4.54)
z z z z
the work W can be described as function of the position vector
y = [ yA , zA , A , z1 , z2 ]T . (4.55)
Due to
W = W (y) (4.56)
principle of virtual work (4.52) leads to
W
W = y = 0 . (4.57)
y
Because of y 6= 0 a system of linear equations in the form of
Ky = b (4.58)
results from (4.57). The matrix K and the vector b are given by

y Q Q
2 cQ 0 2 cQ h0 z Q
c yz cQ



0 2 cR 0 cR cR


Q y Q
K = 2 cQ h0 b
c +h0 yz cQ b (4.59)

0 c 2 R
2 cR h0 z cQ


yQ c cR b Q
c +h0 yz cQ cA + cS + cR cS
z Q 2 R

Q Q
yz cQ cR 2b cR h0 yz cQ cS
cA + cS + cR

57
Vehicle Dynamics FH Regensburg, University of Applied Sciences

and
mA + 2 mR
0



b =
(m1 + m2 ) hR ay .
(4.60)
mR y/z


mR y/z
The following abbreviations have been used:
2  2
y Q

y y b
= + r0 , cA = cA + cQ , c = 2 cQ h20 + 2 cR . (4.61)
z z z z 2

The system of linear equations (4.58) can be solved numerically, e.g. with MATLAB. Thus the
influence of axle suspension and axle kinematics on the roll behavior of the vehicle can be
investigated.

a) A b) A

2 2 0
1 1 0 roll center
roll center

Figure 4.9: Roll Behavior at Cornering: a) without and b) with Camber Compensation

If the wheels only move vertically to the body at bound and rebound, then, at fast cornering
the wheels are no longer perpendicular to the track Fig. 4.9 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.
At more sportive vehicles thus axle kinematics are employed, where the wheels are rotated
around the longitudinal axis at bound and rebound, 1 = 1 (z1 ) and 2 = 2 (z2 ). With this,
a camber compensation can be achieved with 1 0 and 2 0. Fig. 4.9 b. By the rotation
of the wheels around the longitudinal axis on jounce, 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.

4.2.3 Roll Center and Roll Axis

The roll center can be constructed from the lateral motion of the wheel contact points Q1 and
Q2 , Fig. 4.9.
The line through the roll center at the front and rear axle is called roll axis, Fig. 4.10.

58
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

roll axis
roll center front roll center rear

Figure 4.10: Roll Axis

4.2.4 Wheel Loads

The roll angle of a vehicle during cornering depends on the roll stiffness of the axle and on the
position of the roll center.
Different axle layouts at the front and rear axle may result in different roll angles of the front
and rear part of the chassis, Fig. 4.11.

-TT
+TT

PR0+P PR0-P PR0+PR

PR0-PR
PF0+P PF0-P PF0+PF PF0-PF

Figure 4.11: Wheel Loads for a flexible and a rigid Chassis

On most passenger cars the chassis is rather stiff. Hence, front an rear part of the chassis
are forced via an internal torque to an overall chassis roll angle. This torque affects the wheel
loads and generates different wheel load differences at the front and rear axle.
Due to the degressive influence of the wheel load to longitudinal and lateral tire forces the
steering tendency of a vehicle can be affected.

59
Vehicle Dynamics FH Regensburg, University of Applied Sciences

4.2.5 Cornering Resistance

We look at a vehicle cornering with constant velocity v on an flat horizontal road, Fig. 4.12.
Then, the velocity state of the vehicle can be described by

v cos 0
v0S,F = v sin und 0F,F = 0 (4.62)
0
where denotes the side slip angle of the vehicle at the center of gravity, and the subscript ,F
indicates that the vectors are expressed in the vehicle fixed reference frame F .
The angular velocity of the vehicle is given by
v
= , (4.63)
R
where R denotes the radius of curvature.

a2 Fx2
Fy2

a1 S
R
yF
v
xF

Fx1 Fy1

Figure 4.12: Steady State Cornering

In the body fixed reference frame linear and angular momentum result in
v2
 
m sin = Fx1 cos Fy1 sin + Fx2 , (4.64)
R

60
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

v2
 
m cos = Fx1 sin + Fy1 cos + Fy2 , (4.65)
R
0 = a1 (Fx1 sin + Fy1 cos ) a2 Fy2 , (4.66)

where m denotes the mass of the vehicle, Fx1 , Fx2 , Fy1 , Fy2 are the resulting forces in longitu-
dinal and vertical direction applied at the front and rear axle, and specifies the average steer
angle at the front axle.
The engine torque is distributed by the center differential to the front and rear axle. Then, in
steady state condition it holds

Fx1 = k FD und Fx2 = (1 k) FD , (4.67)

where FD is the driving force and by k different driving conditions can be modelled:

k=0 Rear Wheel Drive Fx1 = 0, Fx2 = FD

Fx1 k
0<k<1 All Wheel Drive =
Fx2 1k
k=1 Front Wheel Drive Fx1 = FD , Fx2 = 0

If we insert (4.67) into (4.64) we get

  mv 2
k cos + (1k) FD sin Fy1 = sin ,
R
mv 2 (4.68)
k sin FD + cos Fy1 + Fy2 = cos ,
R
a1 k sin FD + a1 cos Fy1 a2 Fy2 = 0 .

This equations can be resolved for the drive force

a2
cos sin sin cos
a1 + a2 mv 2
FD = . (4.69)
k + (1 k) cos R

The drive force vanishes, if

a2 a2
cos sin = sin cos oder tan = tan (4.70)
a1 + a2 a1 + a2
holds. This corresponds with the Ackermann geometry.
But the Ackermann geometry holds only for small lateral accelerations. In real driving situations
the side slip angle of a vehicle at the center of gravity is always smaller then the Ackermann
side slip angle. Then, due to tan < a1a+a2
2
tan a drive force FD > 0 is needed to overcome
the cornering resistance of the vehicle.

61
Vehicle Dynamics FH Regensburg, University of Applied Sciences

4.3 Simple Handling Model

4.3.1 Forces

The tire forces at the wheel of one axle are combined to one resulting force. Tire torques,
the rolling resistance and aerodynamic forces and torques applied at the vehicle remain unre-
spected.
Unlike with the kinematic tire model, now small lateral motions in the contact points are per-
mitted. At small lateral slips, the lateral force can be approximated by a linear approach

Fy = cS sy (4.71)

where cS is a constant depending on the wheel load Fz and the lateral slip sy is defined by
(2.53).
Because the vehicle is neither accelerated nor decelerated, the rolling condition is fulfilled at
every wheel
rD = eTx v0P . (4.72)
Here rD is the dynamic tire radius, v0P the contact point velocity and ex the unit vector in
longitudinal direction.
With the lateral tire speed
vy = eTy v0P (4.73)
and the rolling condition (4.72) the lateral slip can be calculated from
eTy v0P
sy = , (4.74)
| eTx v0P |
with ey labelling the unit vector in the tires lateral direction.
So the motions of the vehicle model, Fig. 4.13, within the x0 , y0 plane are only influenced by
the lateral forces
Fy1 = cS1 syA1 ; Fy2 = cS2 syA2 . (4.75)

4.3.2 Kinematics

At the single-track vehicle model, as drawn in Fig. 4.13, the yaw motion of the vehicle is de-
scribed by the angle . The vehicle velocity at the center of gravity is given as constant

v = const. (4.76)

With the side slip angle the vehicle velocity can be expressed in the vehicle fixed frame xF ,
yF , zF
v cos
vSF ,F = v sin . (4.77)
0

62
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

y0

1
a1

x
1
y
yF A1

a2 Fy1

v
xF

y2
SF

A2 x2

Fy2

x0

Figure 4.13: Single Track Vehicle Model

For the calculation of the lateral slips, the speed vectors and the unit vectors in longitudinal
and lateral direction of the single axles are needed. One gets

cos sin v cos
ex1 ,F = sin , ey1 ,F = cos , v0A1 ,F = v sin + a1 (4.78)
0 0 0
and
1 0 v cos
ex2 ,F = 0 , ey2 ,F = 1 , v0A2 ,F = v sin a2 . (4.79)
0 0 0

4.3.3 Lateral Slips

With (4.79), the lateral slip at the front axle follows from (4.74):

+ sin (v cos ) cos (v sin + a1 )

syA1 = . (4.80)
| cos (v cos ) + sin (v sin + a1 ) |
The lateral slip at the rear axle is given by
v sin a2
syA2 = . (4.81)
| v cos |

The yaw velocity of the vehicle , the side slip angle and the steering angle are considered
to be small
| a1 |  |v| ; | a2 |  |v| (4.82)

63
Vehicle Dynamics FH Regensburg, University of Applied Sciences

| |  1 and | |  1 . (4.83)
Because the side slip angle always labels the smaller angle between speed vector and vehicle
longitudinal axis, instead of v sin v the approximation

v sin |v| (4.84)

has to be used.
Respecting (4.82), (4.83) and (4.84), from (4.80) and (4.81) then follow
a1 v
syA1 = + (4.85)
|v| |v|
and
a2
syA2 = + . (4.86)
|v|

4.3.4 Equations of Motion

To derive the equations of motion, the velocities, angular velocities and the accelerations are
needed.
For small side slip angles  1, (4.77) can be approximated by

v
vSF ,F = |v| . (4.87)
0

The angular velocity is given by

0
0F,F = 0 . (4.88)

If the vehicle accelerations are also expressed in the vehicle fixed frame xF , yF , zF , one finds
at constant vehicle speed v = const and with neglecting small higher order terms

0
aSF ,F = 0F,F vSF ,F + vSF ,F = v + |v| . (4.89)
0

The angular acceleration is given by

0
0F,F = 0 (4.90)

where the substitution

= (4.91)

64
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

was used. The linear momentum in the vehicles lateral direction reads as
m (v + |v| ) = Fy1 + Fy2 , (4.92)
where, due to the small steering angle, the term Fy1 cos has been approximated by Fy1 and
m describes the vehicle mass.
With (4.91) the angular momentum delivers
= a1 Fy1 a2 Fy2 , (4.93)
where names the inertia of vehicle around the vertical axis.
With the linear description of the lateral forces (4.75) and the lateral slips (4.85), (4.86) one
gets from (4.92) and (4.93) two coupled, but linear first order differential equations
a1 v a2
   
cS1 cS2 v
= + + + , (4.94)
m |v| |v| |v| m |v| |v| |v|
a1 v a2
   
a1 cS1 a2 cS2
= + + , (4.95)
|v| |v| |v|
which can be written in the form of a state equation

cS1 + cS2 a2 cS2 a1 cS1 v   v cS1
 
m |v| m |v||v| + |v| m |v|
|v|
 
= . (4.96)
|{z}
a2 cS2 a1 cS1 a21 cS1 + a22 cS2 | {z } v a1 cS1
u
| {z } x
x |v| |v|
| {z } | {z }
A B
If a system can be, at least approximatively, described by a linear state equation, then, stabil-
ity, steady state solutions, transient response, and optimal controlling can be calculated with
classic methods of system dynamics.

4.3.5 Stability
4.3.5.1 Eigenvalues

The homogeneous state equation

x = A x (4.97)
describes the eigen-dynamics. If the approach
xh (t) = x0 et (4.98)
is inserted into (4.97), then the homogeneous equation remains
( E A) x0 = 0 . (4.99)
Non-trivial solutions x0 6= 0 one gets for
det | E A| = 0 . (4.100)
The eigenvalues provide information about the stability of the system.

65
Vehicle Dynamics FH Regensburg, University of Applied Sciences

4.3.5.2 Low Speed Approximation

The state matrix

cS1 + cS2 a2 cS2 a1 cS1 v

m |v| m |v||v| |v|
Av0 =


(4.101)
2 2
a1 cS1 + a2 cS2
0
|v|
approximates at v 0 the eigen-dynamics of vehicles at low speeds.
The matrix (4.101) has the eigenvalues

cS1 + cS2 a21 cS1 + a22 cS2

1v0 = and 2v0 = . (4.102)
m |v| |v|
The eigenvalues are real and, independent from the driving direction, always negative.
Thus, vehicles at low speed possess an asymptotically stable driving behavior!

4.3.5.3 High Speed Approximation

At highest driving velocities v , the state matrix can be approximated by

v
0
Av

=
|v|
. (4.103)
a2 cS2 a1 cS1
0

Using (4.103) one receives from (4.100) the relation
v a2 cS2 a1 cS1
2v + = 0 (4.104)
|v|
with the solutions r
v a2 cS2 a1 cS1
1,2v = . (4.105)
|v|
When driving forward with v > 0, the root argument is positive, if

a2 cS2 a1 cS1 < 0 (4.106)

holds. Then however, one eigenvalue is positive and the system is unstable.
Two zero-eigenvalues 1 = 0 and 2 = 0 one gets for

a1 cS1 = a2 cS2 . (4.107)

The driving behavior is then indifferent. Slight parameter variations however can lead to an
unstable behavior.

66
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

With
a2 cS2 a1 cS1 > 0 or a1 cS1 < a2 cS2 (4.108)
and v > 0 the root argument in (4.105) becomes negative. The eigenvalues are then imagi-
nary, and disturbances lead to undamped vibrations.
To avoid instability, high-speed vehicles have to satisfy the condition (4.108).
The root argument in (4.105) changes at backward driving its sign. A vehicle showing stable
driving behavior at forward driving becomes unstable at fast backward driving!

4.3.6 Steady State Solution

4.3.6.1 Side Slip Angle and Yaw Velocity

With a given steering angle = 0 , after a certain time, a stable system reaches steady state.
With xst = const. or xst = 0, the state equation (4.96) is reduced to a linear system of equations

A xst = B u . (4.109)

With the elements from the state matrix A and the vector B one gets from (4.109) two equa-
tions to determine the steady state side slip angle st and the steady state angular velocity st
at a constant given steering angle = 0

|v| (cS1 + cS2 ) st + (m v |v| + a1 cS1 a2 cS2 ) st = v cS1 0 , (4.110)

|v| (a1 cS1 a2 cS2 ) st + (a21 cS1 + a22 cS2 ) st = v a1 cS1 0 , (4.111)
where the first equation has been multiplied by m |v| |v| and the second with |v|. The
solution can be derived from

v cS1 0 m v |v| + a1 cS1 a2 cS2

v a1 cS1 0 a21 cS1 + a22 cS2
st = (4.112)
|v| (cS1 + cS2 ) m v |v| + a1 cS1 a2 cS2
|v| (a1 cS1 a2 cS2 ) a21 cS1 + a22 cS2

and
|v| (cS1 + cS2 ) v cS1 0
|v| (a1 cS1 a2 cS2 ) v a1 cS1 0
st = (4.113)
|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 ) . (4.114)

67
Vehicle Dynamics FH Regensburg, University of Applied Sciences

For a non vanishing denominator detD 6= 0 steady state solutions exist

a1
a2 m v |v|
v cS2 (a1 + a2 )
st = 0 , (4.115)
|v| a + a + m v |v| a2 cS2 a1 cS1
1 2
cS1 cS2 (a1 + a2 )
v
st = a2 cS2 a1 cS1 0 . (4.116)
a1 + a2 + m v |v|
cS1 cS2 (a1 + a2 )
At forward driving vehicles v > 0 the steady state side slip angle, starts with the kinematic
value
v0 v a2 v0 v
st = 0 and st = 0 (4.117)
|v| a1 + a2 a1 + a2
and decreases with increasing speed. At speeds larger then
s
a2 cS2 (a1 + a2 )
vst=0 = (4.118)
a1 m

the side slip angle changes the sign.

Using the kinematic value of the yaw velocity equation (4.116) can be written as

v 1
st = 0 , (4.119)
a1 + a2  
2
|v| v
1 +
v vch

where s
cS1 cS2 (a1 + a2 )2
vch = (4.120)
m (a2 cS2 a1 cS1 )
is called the characteristic speed of the vehicle.
Because the rear wheels are not steered, higher slip angles at the rear axle can only be
reached by slanting the car.
In Fig. 4.14 the side slip angle , and the driven curve radius R are plotted versus the driving
speed v . The steering angle has been set to 0 = 1.4321 , in order to let the vehicle drive
a circle with the radius R0 = 100 m at v 0. The actually driven circle radius R has been
calculated via
v
st = . (4.121)
R
Some concepts for an additional steering of the rear axle were trying to keep the vehicles
side slip angle to zero by an appropriate steering or controlling. Due to numerous problems
production stage could not yet be reached.

68
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

steady state side slip angle radius of curvrature

2 200

0
150
-2
[deg]

r [m]
-4 100

-6 a1*c S1/a2*c S2 = 0.66667 a1*cS1/a2*cS2 = 0.66667

50
a1*cS1/a2*c S2 = 1 a1*cS1/a2*cS2 = 1
-8
a1*cS1/a2*c S2 = 1.3333 a1*cS1/a2*cS2 = 1.3333
-10 0
0 10 20 30 40 0 10 20 30 40
v [m/s] v [m/s]
110 770 N m
m=700 kg; a1 =1.2 m;
cS1 = 80 000 N m; cS2 = 73 846 N m
=1000 kg m2 ; a2 =1.3 m;
55 385 N m
Figure 4.14: Steady State Cornering

4.3.6.2 Steering Tendency

After reaching the steady state solution, the vehicle moves in a circle. When inserting (4.121)
into (4.116) and resolving for the steering angle, one gets

a1 + a2 v 2 v a2 cS2 a1 cS1
0 = + m . (4.122)
R R |v| cS1 cS2 (a1 + a2 )

The first term is the Ackermann steering angle, which follows from (4.2) 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,
because at v 0 the second expression in (4.122) becomes neglectably small.
At higher speeds, depending on the value of a2 cS2 a1 cS1 and the driving direction (forward:
v > 0, backward: v < 0), the necessary steering angle differs from the Ackermann-steering
angle. The difference is proportional to the lateral acceleration

v2
ay = . (4.123)
R
At v > 0 the steering tendency of a vehicle is defined by the position of the center of gravity
a1 , a2 and the cornering stiffnesses at the axles cS1 , cS2 .
The various steering tendencies are arranged in the table 4.1.

69
Vehicle Dynamics FH Regensburg, University of Applied Sciences

a1 cS1
understeer 0 > 0A or a1 cS1 < a2 cS2 or <1
a2 cS2
a1 cS1
neutral 0 = 0A or a1 cS1 = a2 cS2 or =1
a2 cS2
a1 cS1
oversteer 0 < 0A or a1 cS1 > a2 cS2 or >1
a2 cS2
Table 4.1: Steering Tendency of a Vehicle at Forward Driving

4.3.6.3 Slip Angles

With the conditions for a steady state solution st = 0, st = 0 and the relation (4.121), the
equations of motion (4.92) and (4.93) can be dissolved for the lateral forces

a2 v2
Fy1st = m , a1 Fy2st
a1 + a2 R or = . (4.124)
a2 Fy1st
a1 v2
Fy2st = m
a1 + a2 R
With the linear tire model (4.71) one gets

Fy1st = cS1 sst

yA1 and Fy2st = cS2 sst
yA2 , (4.125)

where sst st
yA1 and syA2 label the steady state lateral slips at the axles. From (4.124) and (4.125)
now follows
a1 Fy2st cS2 sst
yA2 a1 cS1 sst
y
= = or = stA2 . (4.126)
a2 Fy1st cS1 sst
yA1 a2 cS2 syA1
That means, at a vehicle with understeer tendency (a1 cS1 < a2 cS2 ) during steady state corner-
ing the slip angles at the front axle are larger then the slip angles at the rear axle, sst st
yA1 > syA2 .
So, the steering tendency can also be determined from the slip angle at the axles.

4.3.7 Influence of Wheel Load on Cornering Stiffness

With identical tires at the front and rear axle, given a linear influence of wheel load on the raise
of the lateral force over the lateral slip,

clin lin
S1 = cS Fz1 and cS2 = cS Fz2 . (4.127)

holds. The weight of the vehicle G = m g is distributed over the axles according to the position
of the center of gravity
a2 a1
Fz1 = G and .Fz2 = G (4.128)
a1 + a2 a1 + a2

70
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

With (4.127) and (4.128) one gets

a2
a1 clin
S1 = a1 cS G (4.129)
a1 + a2
and
a1
a2 clin
S2 = a2 cS G. (4.130)
a1 + a2
A vehicle with identical tires would thus be steering neutrally at a linear influence of wheel load
on the slip stiffness, because of
a1 clin lin
S1 = a2 cS2 (4.131)
The fact that the lateral force is applied behind the center of the contact area at the caster
v v
offset distance, leads, because of a1 a1 |v| nL1 and a2 a2 + |v| nL1 to a stabilization of
the driving behavior, independent from the driving direction.
At a real tire, a digressive influence of wheel load on the tire forces is observed, Fig. 4.15.

5 Fz [N ] Fy [N ]
0 0
4 1000 758
2000 1438
Fy [kN]

3
3000 2043
2 4000 2576
5000 3039
1 6000 3434
7000 3762
0
0 1 2 3 4 5 6 7 8 8000 4025
Fz [kN]

Figure 4.15: Lateral Force Fy over Wheel Load Fz at different Slip Angles

According to (4.93) 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 (4.132)
holds.
At a vehicle with the wheel base a = 2.45 m the axle loads Fz1 = 4000 N and Fz2 = 3000 N
deliver 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 receives from the table in 4.15 Fy1 = 2576 N and Fy2 = 2043 N . With this,
the condition (4.132) delivers 1.05 2576 1.45 2043 = 257.55 . The value is significantly
negative and thus stabilizing.
Vehicles with a1 < a2 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 ), a stable driving
behavior can generally only be achieved with different tires.

71
Vehicle Dynamics FH Regensburg, University of Applied Sciences

At increasing lateral acceleration the vehicle is more and more supported by the outer wheels.
At a sufficiently rigid vehicle body the wheel load differences can differ, because of different
kinematics (roll support) or different roll stiffnesses
Due to the digressive influence of wheel load, the deliverable lateral force at an axle decreases
with increasing wheel load difference.
If the wheel load is split more strongly at the front axle than at the rear axle, the lateral force
potential at the front axle decreases more than at the rear axle and the vehicle becomes more
stable with increasing lateral force, i.e. more understeering.

72
5 Vertical Dynamics

5.1 Goals

The aim of vertical dynamics is the tuning of body suspension and damping to guarantee good
driving comfort, resp. a minimal stress of the load at sufficient 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 Fx and lateral force Fy by the coefficient of
friction. The digressive influence of Fz on Fx and Fy as well as instationary processes at the
increase of Fx and Fy 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 wheel load. Small
variations of wheel load also reduce the stress on the track.
The comfort of a vehicle is subjectively judged by the driver. In literature, different approaches
of describing the human sense of vibrations by different metrics can be found.
Transferred to vehicle vertical dynamics, the driver primarily registers the amplitudes and ac-
celerations of the body vibrations. These values are thus used as objective criteria in practice.

5.2 Basic Tuning

5.2.1 Simple Models

Fig. 5.1 shows simple quarter car models, that are suitable for basic investigations of body and
axle vibrations.
At normal vehicles the wheel mass m is in relation to the respective body mass M much
smaller m  M . The coupling of wheel and body movement can thus be neglected for basic
investigations.
In describing the vertical movements of the body, the wheel movements remain unrespected.
If the wheel movements are in the foreground, then body movements can be neglected.
The equations of motion for the models read as

M zB + dS zB + cS zB = dS zR + cS zR (5.1)

73
Vehicle Dynamics FH Regensburg, University of Applied Sciences

``
zB cS `` dS
M 6 ` `

` m zW
6
cS ``
` dS
``
``
`` c
zR
6 `` T
c zR
6
c

Figure 5.1: Simple Vehicle and Suspension Model

and
m zW + dS zW + (cS + cT ) zW = cT zR , (5.2)
where zB and zW label the vertical movements of the body and the wheel mass out of the
equilibrium position. The constants cS , dS describe the body suspension and damping, and
cT the vertical stiffness of the tire. The tire damping is hereby neglected against the body
damping.

5.2.2 Track

The track is given as function in the space domain

zR = zR (x) . (5.3)

In (5.1) also the time gradient of the track irregularities is necessary. From (5.3) firstly follows

d zR dx
zR = . (5.4)
dx dt
At the simple model the speed, with which the track irregularities are probed equals the vehicle
speed dx/dt = v . If the vehicle speed is given as time function v = v(t), the covered distance
x can be calculated by simple integration.

5.2.3 Spring Preload

The suspension spring is loaded with the respective vehicle load. At linear spring characteris-
tics the steady state spring deflection is calculated from

Mg
f0 = . (5.5)
cS

74
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

At a conventional suspension without niveau regulation a load variation M M + 4M leads

to changed spring deflections f0 f0 + 4f . In analogy to (5.5) the additional deflection
follows from
4M g
4f = . (5.6)
cS
If for the maximum load variation 4M max the additional spring deflection is limited to 4f max
the suspension spring rate can be estimated by a lower bound
4M max g
cS . (5.7)
4f max

5.2.4 Eigenvalues

At an ideally even track the right side of the equations of motion (5.1), (5.2) vanishes because
of zR = 0 and zR = 0. The remaining homogeneous second order differential equations can be
written as
z + 2 z + 02 z = 0 . (5.8)
The respective attenuation constants and the undamped natural circular frequency 0 for
the models in Fig. 5.1 can be determined from a comparison of (5.8) with (5.1) and (5.2). The
results are arranged in table 5.1.

attenuation undamped
Motions Differential Equation
constant Eigenfrequency

dS cS
Body M zB + dS zB + cS zB = 0 B = B2 0 =
2M M
dS 2 cS + cT
Wheel m zW + dS zW + (cS + cT ) zW = 0 R = W =
2m 0
m
Table 5.1: Attenuation Constants and undamped natural Frequencies

With
z = z0 et (5.9)
the equation
(2 + 2 + 02 ) z0 et = 0 . (5.10)
follows from (5.8). For
2 + 2 + 02 = 0 (5.11)
also non-trivial solutions are possible. The characteristical equation (5.11) has got the solutions
q
1,2 = 2 02 (5.12)

75
Vehicle Dynamics FH Regensburg, University of Applied Sciences

For 2 02 the eigenvalues 1,2 are real and, because of 0 not positive, 1,2 0.
Disturbances z(t = 0) = z0 with z(t = 0) = 0 then subside exponentially.
With 2 < 02 the eigenvalues become complex
q
1,2 = i 02 2 . (5.13)

The system now executes damped oscillations.

The case
2 = 02 , bzw. = 0 (5.14)
describes, in the sense of stability, an optimal system behavior.
Wheel and body mass, as well as tire stiffness are fixed. The body spring rate can be calcu-
lated via load variations, cf. section 5.2.3. With the abbreviations from table 5.1 now damping
parameters can be calculated from (5.14) which provide with
r
cS p
(dS )opt1 = 2 M = 2 cS M (5.15)
M
optimal body vibrations and with
r
cS + cT p
(dS )opt2 = 2 m = 2 (cS + cT ) m (5.16)
m
optimal wheel vibrations.

5.2.5 Free Vibrations

Fig. 5.2 shows the time response of a damped single-mass oscillator to an initial disturbance
as results from the solution of the differential equation (5.8). The system here has been started
without initial speed z(t = 0) = 0 but with the initial disturbance z(t = 0) = z0 . If the attenuation
constant is increased at first the system approaches the steady state position zG = 0 faster
and faster, but then, a slow asymptotic behavior occurs.
Counting differences from the steady state positions as errors (t) = z(t) zG , allows judging
the quality of the vibration. The overall error is calculated by
t=t
Z E
2G = z(t)2 dt , (5.17)
t=0

where the time tE have to be chosen appropriately. If the overall error becomes a Minimum

2G M inimum (5.18)

the system approaches the steady state position as fast as possible.

76
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

z0

z(t)

Figure 5.2: Damped Vibration

To judge driving comfort and safety the deflections zB and accelerations zB of the body and
the dynamic wheel load variations are used.
The system behavior is optimal if the parameters M , m, cS , dS , cT result from the demands
for comfort
t=t
Z E n
2
2 o
2GC = g1 zB + g2 zB dt M inimum (5.19)
t=0

and safety
t=t
Z E

2
2GS = cT zW dt M inimum . (5.20)
t=0

With the factors g1 and g2 deflections and accelerations can be weighted differently. In the
equations of motion for the body (5.1) the terms M zB and cS zB are added. With g1 = M and
g2 = cS one gets system-fitted weighting factors.
At the damped single-mass oscillator, the integrals in (5.19) can, for tE , still be solved
analytically. One gets  
cS 1 dS cS
2GC = zB2 0 + 2 (5.21)
M 2 M dS
and  
1 dS m
2GS = 2
zW c2T + . (5.22)
0
2 cS + cT dS
Small body suspension stiffnesses cS 0 or large body masses M make the comfort
criteria (5.21) small 2GC 0 and so guarantee a high driving comfort.
A great body mass however is uneconomic. The body suspension stiffness cannot be reduced
arbitrary low values, because then load variations would lead to too great changes in static
deflection. At fixed values for cS and M the damper can be designed in a way that minimizes
the comfort criteria (5.21). From the necessary condition for a minimum

2GC
 
cS 1 1 cS
= zB2 0 2 2 = 0 (5.23)
dS M 2 M dS

77
Vehicle Dynamics FH Regensburg, University of Applied Sciences

the optimal damper parameter p

(dS )opt3 = 2 cS M , (5.24)
that guarantees optimal comfort follows.
Small tire spring stiffnesses cT 0 make the safety criteria (5.22) small 2GS 0 and thus
reduce dynamic wheel load variations. The tire spring stiffness can however not be reduced to
arbitrary low values, because this would cause too great tire deformation. Small wheel masses
m 0 and/or a hard body suspension cS also reduce the safety criteria (5.22). The use
of light metal rims increases, because of wheel weight reduction, the driving safety of a car.
Hard body suspensions contradict driving comfort.
With fixed values for cS , cT and m here the damper can also be designed to minimize the
safety criteria (5.22). From the necessary condition of a minimum
2GS
 
2 1 1 m
= zW c2
0 T
2 = 0 (5.25)
dS 2 cS + cT dS
the optimal damper parameter
p
(dS )opt4 = (cS + cT ) m , (5.26)

follows, which guarantees optimal safety.

5.3 Nonlinear Force Elements

5.3.1 Quarter Car Model

The principal influence of nonlinear characteristics on driving comfort and safety can already
be displayed on a quarter car model Fig. 5.3.

zB
progressive spring degressive damper
M
FF FD

FR
xR v

x zW
m

cT zR

Figure 5.3: Quarter Car Model with nonlinear Characteristics

The equations of motion are given by

M zB = F M g
(5.27)
m zW = Fz F m g ,

78
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

where g = 9.81 m/s2 labels the constant of gravity and M , m are the masses of body and
wheel. The coordinates zB and zW are measured from the equilibrium position.
Thus, the wheel load Fz is calculated from the tire deflection zW zR via the tire stiffness cT
Fz = (M + m) g + cT (zR zW ) . (5.28)
The first term in (5.28) describes the static part. The condition Fz 0 takes the wheel lift off
into consideration.
Body suspension and damping are described with nonlinear functions of the spring travel
x = zW zB (5.29)
and the spring velocity
v = zW zB , (5.30)
where x > 0 and v > 0 marks the spring and damper compression.
The damper characteristics are modelled as digressive functions with the parameters pi 0,
i = 1(1)4
1
p1 v 1 + p2 v v 0 (Druck)



FD (v) = . (5.31)
1
p3 v v < 0 (Zug)


1 p4 v
A linear damper with the constant d is described by p1 = p3 = d and p2 = p4 = 0.
For the spring characteristics the approach
FR 1 p5
FF (x) = M g + x (5.32)
xR |x|
1 p5
xR
is used, where M g marks the spring preload. With parameters within the range 0 p5 < 1,
one gets differently progressive characteristics. The special case p5 = 0 describes a linear
spring with the constant c = FR /xR . All spring characteristics run through the operating point
xR , FR . Thus, at a real vehicle, one gets the same roll angle, independent from the chosen
progression at a certain lateral acceleration.

5.3.2 Random Road Profile

The vehicle moves with the constant speed vF = const. When starting at t = 0 at the point
xF = 0, the current position of the car is given by
xF (t) = vF t . (5.33)
The irregularities of the track can thus be written as time function zR = zR (xF (t))
The calculation of optimal characteristics, i.e. the determination of the parameters p1 to p5 ,
is done for three different tracks. Each track consists of a number of single obstacles, which
lengths and heights are distributed randomly. Fig. 5.4 shows the first track profile zS1 (x). Pro-
files number two and three are generated from the first by multiplication with the factors 3 and
5, zS2 (x) = 3 zS1 (x), zS3 (x) = 5 zS1 (x).

79
Vehicle Dynamics FH Regensburg, University of Applied Sciences

road profil [m]

0.1

0.05

-0.05

-0.1
0 20 40 60 [m] 80 100

Figure 5.4: Track profile 1

5.3.3 Vehicle Data

The values, arranged in table 5.2, describe the respective body mass of a fully loaded and an
empty bus over the rear axle, the mass of the rear axle and the sum of tire stiffnesses at the
twin tire rear axle.

vehicle data M [kg] m [kg] FR [N] xR [m] cT [N/m]

fully loaded 11 000 800 40 000 0.100 3 200 000
unloaded 6 000 800 22 500 0.100 3 200 000

Table 5.2: Vehicle Data

The vehicle possesses niveau-regulation. Therefore also the force FR at the reference deflec-
tion xR has been fitted to the load.
The vehicle drives at the constant speed vF = 20 m/s.
The five parameters, pi , i = 1(1)5, which describe the nonlinear spring-damper characteristics,
are calculated by minimizing quality functions.

5.3.4 Quality Criteria

In a first quality function, driving comfort and safety are to be judged by body accelerations
and wheel load variations
Z tE   2  F D 2 
1 zB z
GK1 = + . (5.34)
tE t0 t0 g FzS
| {z } | {z }
comfort safety

The body acceleration zB has been normalized to the constant of gravity g . The dynamic
share of the normal force FzD = cT (zR zW ) follows from (5.28) with the static normal force
FzS = (M + m) g .

80
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

At real cars the spring travel is limited. The quality criteria is therefore extended accordingly
Z tE   2  P 2  x 2 
1 zB D
GK2 = + + , (5.35)
tE t0 t0 g PS xR
| {z } | {z } | {z }
comfort safety spring travel

where the spring travel x, defined by (5.29), has been related to the reference travel xr .
According to the covered distance and chosen driving speed, the times used in (5.34) and
(5.35) have been set to t0 = 0 s and tE = 8 s

5.3.5 Optimal Parameter

5.3.5.1 Linear Characteristics

Judging the driving comfort and safety after the criteria GK1 and restricting to linear character-
istics, with p1 = p3 and p2 = p4 = p5 = 0, one gets the results arrayed in table5.3. The spring

optimal parameter parts in quality criteria

road load p1 p2 p3 p4 p5 comfort safety
1 + 35766 0 35766 0 0 0.002886 0.002669
2 + 35763 0 35763 0 0 0.025972 0.024013
3 + 35762 0 35762 0 0 0.072143 0.066701
1 20298 0 20298 0 0 0.003321 0.003961
2 20300 0 20300 0 0 0.029889 0.035641
3 19974 0 19974 0 0 0.083040 0.098385

Table 5.3: Linear Spring and Damper Parameter optimized via GK1

constants c = FR /xr for the fully loaded and the empty vehicle are defined by the numerical
values in table 5.2. One gets:cempty = 225 000N/m and cloaded = 400 000N/m.
As expected the results are almost independent from the track. The optimal value of the damp-
ing parameter d = p1 = p3 however is strongly dependent on the load state. The optimizing
quasi fits the damper constant to the changed spring rate.
The loaded vehicle is more comfortable and safer.

5.3.5.2 Nonlinear Characteristics

The results of the optimization with nonlinear characteristics are arrayed in the table 5.4.
The optimizing has been started with the linear parameters from table 5.3. Only at the extreme
track irregularities of profile 3, linear spring characteristics, with p5 = 0, appear, Fig. 5.6. At
moderate track irregularities, one gets strongly progressive springs.

81
Vehicle Dynamics FH Regensburg, University of Applied Sciences

optimal parameter parts in quality criteria

road load p1 p2 p3 p4 p5 comfort safety
1 + 16182 0.000 20028 1.316 0.9671 0.000265 0.001104
2 + 52170 2.689 57892 1.175 0.6983 0.009060 0.012764
3 + 1875 3.048 311773 4.295 0.0000 0.040813 0.050069
1 13961 0.000 17255 0.337 0.9203 0.000819 0.003414
2 16081 0.808 27703 0.454 0.6567 0.012947 0.031285
3 9942 0.227 64345 0.714 0.0000 0.060992 0.090250

Table 5.4: Nonlinear Spring and Damper Characteristics optimized via GK1

The dampers are digressive and differ in jounce and rebound.

In comparison to the linear model a significant improvement can be noted, especially in com-
fort.
While driving over profile 2 with the loaded vehicle, the body accelerations are displayed in
Fig. 5.5.

body accelerations [m/s2 ]

10

-5

-10
0 2 4 [s] 6 8

Figure 5.5: Body Accelerations optimized via GK1 ( linear, nonlinear)

spring force [kN]

40

20

-20

-40
-0.1 -0.05 0 0.05 0.1
spring travel [m]

Figure 5.6: Optimal Spring Characteristics for fully loaded Vehicle; Criteria: GK1

82
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

The extremely progressive spring characteristics, optimal at smooth tracks (profile 1), cannot
be realize practically in that way. Due to the small spring stiffness around the equilibrium posi-
tion, small disturbances cause only small aligning forces. Therefore it would take long to reach
the equilibrium position again. Additionally, friction forces in the body suspension would cause
a large deviation of the equilibrium position.

5.3.5.3 Limited Spring Travel

Practically relevant results can only be achieved, if additionally the spring travels are judged.
Firstly, linear characteristics are assumed again, table 5.5.

optimal parameter parts in quality criteria

road load p1 p2 p3 p4 p5 comfort safety s. travel
1 + 68727 0 68727 0 0 0.003854 0.003673 0.006339
2 + 68666 0 68666 0 0 0.034657 0.033025 0.057097
3 + 72882 0 72882 0 0 0.098961 0.094431 0.148757
1 35332 0 35332 0 0 0.004417 0.004701 0.006638
2 35656 0 35656 0 0 0.040049 0.042507 0.059162
3 37480 0 37480 0 0 0.112143 0.116722 0.155290

Table 5.5: Linear Spring and Damper Characteristics optimized via GK2

The judging numbers for comfort and safety have worsened by limiting the spring travel in
comparison to the values from table 5.3.
In order to receive realistic spring characteristics, now the parameter p5 has been limited up-
wards to p5 0.6. Starting with the linear parameters from table 5.5, an optimization via

optimal parameter parts in quality criteria

road load p1 p2 p3 p4 p5 comfort safety s. travel
1 + 175530 12.89 102997 3.437 0.4722 0.001747 0.002044 0.005769
2 + 204674 5.505 107498 1.234 0.6000 0.015877 0.018500 0.050073
3 + 327864 4.844 152732 1.165 0.5140 0.064980 0.068329 0.116555
1 66391 5.244 50353 2.082 0.5841 0.002380 0.003943 0.005597
2 37246 0.601 37392 0.101 0.5459 0.024524 0.033156 0.059717
3 89007 1.668 68917 0.643 0.3614 0.085001 0.102876 0.125042

Table 5.6: Nonlinear Spring and Damper Characteristics optimized via GK2

criteria GK2 delivers the results arranged in table 5.6.

A vehicle with GK2 -optimized characteristics manages the travel over uneven tracks with sig-
nificantly less spring travel than a vehicle with GK1 -optimized characteristics, Fig. 5.7.

83
Vehicle Dynamics FH Regensburg, University of Applied Sciences

spring travel [m]

0.1

0.05

-0.05

-0.1
0 2 4 [s] 6 8

Figure 5.7: Spring Travels on Profile 2 (- - - GK1 , GK2 )

The reduced spring travel however reduces comfort and safety.

Still, in most cases, the according part of the quality criteria in table 5.6 lie even below the
values of the linear model from table 5.3, where the spring travels have not been evaluated.
By the use of nonlinear characteristics, the comfort and safety of a vehicle can so be improved,
despite limitation of the spring travel.
The optimal damper characteristics strongly depend on the roughness of the track, Fig. 5.8.

damper force [kN]

100
rebound
50

-50
compression
-100
-1 -0.5 0 [m/s] 0.5 1

Figure 5.8: Optimal Damper Characteristics according to Table 5.6

Optimal comfort and safety are only guaranteed if the dampers are fitted to the load as well as
to the roughness of the track.

5.4 Dynamic Force Elements

5.4.1 System Response in the Frequency Domain

5.4.1.1 First Harmonic Oscillation

The effect of dynamic force elements is usually judged in the frequency domain. For this, on
test rigs or in simulation, the force element is periodically excited with different frequencies

84
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

f0 fi fE and amplitudes Amin Aj Amax

xe (t) = Aj sin(2 fi t) . (5.36)
Starting at t = 0 at t = T0 with T0 = 1/f0 the system usually is in a steady state condition. Due
to the nonlinear system behavior the system response is periodic, yet not harmonic. For the
evaluation thus the answer, e.g. the measured or calculated force F , each within the intervals
tSi t tSi +Ti , is approximated by harmonic functions as good as possible
F (t) i sin(2 fi t) + i cos(2 fi t) . (5.37)
|{z} | {z }
measured f irst harmonic approximation
or
calculated

The coefficients i and i can be calculated from the demand for a minimal overall error
tSi+Ti
Z 
1 2
i sin(2 fi t)+i cos(2 fi t) F (t) dt M inimum . (5.38)
2
tSi

The differentiation of (5.38) with respect to i and i delivers two linear equations as necessary
conditions
tSi+Ti
Z  2
i sin(2 fi t)+i cos(2 fi t) F (t) sin(2 fi t) dt = 0
tSi
tSi+Ti (5.39)
Z  2
i sin(2 fi t)+i cos(2 fi t) F (t) cos(2 fi t) dt = 0
tSi

with the solutions

R R R R
F sin dt cos2 dt F cos dt sin cos dt
i =
sin2 dt cos2 dt 2 sin cos dt
R R R
, (5.40)
F cos dt sin2 dt F sin dt sin cos dt
R R R R
i =
sin2 dt cos2 dt 2 sin cos dt
R R R

where the integral limits and arguments of sine and cosine have no longer been written.
Because it is integrated exactly over one period tSi t tSi +Ti , for the integrals in (5.40)
Ti R Ti
sin2 dt =
R R
sin cos dt = 0 ; ; cos2 dt = (5.41)
2 2
holds, and as solution
Z Z
2 2
i = F sin dt , i = F cos dt . (5.42)
Ti Ti
remains. These however are exactly the first two coefficients of a FourierApproximation.
In practice, the frequency response of a system is not determined punctual, but continuous.
For this, the system is excited by a sweep-sine.

85
Vehicle Dynamics FH Regensburg, University of Applied Sciences

5.4.1.2 Sweep-Sine Excitation

In analogy to the simple sine-function

xe (t) = A sin(2 f t) , (5.43)

where the period duration T = 1/f appears as pre-factor at differentiation

2
xe (t) = A 2 f cos(2 f t) = A cos(2 f t) , (5.44)
T
now a generalized sine-function can be constructed. Starting with

xe (t) = A sin(2 h(t)) (5.45)

the time derivative results in

xe (t) = A 2 h(t) cos(2 h(t)) . (5.46)

Now we demand, that the function h(t) delivers a period, that fades linear in time, i.e:
1 1
h(t) = = , (5.47)
T (t) pqt
where p > 0 and q > 0 are constants yet to determine. From (5.47)
1
h(t) = ln(p q t) + C (5.48)
q
follows. The initial condition h(t = 0) = 0 fixes the integration constant
1
C = ln p . (5.49)
q
Inserting (5.49) in (5.48), a sine-like function follows from (5.45)
 2 p 
xe (t) = A sin ln , (5.50)
q pqt
delivering linear fading period durations.
The important zero values for determining the period duration lie at
1 p p
ln = 0, 1, 2, or = en q , mit n = 0, 1, 2, (5.51)
q p q tn p q tn
and
p
tn = (1 en q ) , n = 0, 1, 2, . (5.52)
q
The time difference between two zero points determines the period
p
Tn = tn+1 tn = (1e(n+1) q 1+en q )
q
p n q , n = 0, 1, 2, . (5.53)
Tn = e (1 eq )
q

86
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

For the first (n = 0) and last (n = N ) period one finds

p
T0 = (1 eq )
q
p . (5.54)
TN = (1 eq ) eN q = T0 eN q
q
With the frequency range to investigate given by the initial f0 and final fE frequency, the pa-
rameters q and the relation q/p can be calculated from (5.54)
1
1 fE q n hf iN o
E
q = ln , = f0 1 , (5.55)
N f0 p f0
with N fixing the number of frequency intervals. The passing of the whole frequency range
then takes
1 e(N +1) q
tN +1 = (5.56)
q/p
seconds.

5.4.2 Hydro-Mount

5.4.2.1 Principle and Model

For elastic suspension of engines in vehicles very often specially developed hydro-mounts
are used. The dynamic nonlinear behavior of these components guarantees a good acoustic
decoupling, but simultaneously provides sufficient damping.

xe
main spring

chamber 1
cT cF
__ c__
T
membrane 2 uF 2

ring channel MF
dF
__ dF
__
2 2
chamber 2

Figure 5.9: Hydro-Mount

87
Vehicle Dynamics FH Regensburg, University of Applied Sciences

Fig. 5.9 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 relation of the chamber cross section to ring
channel cross section is very large. Thus the fluid is moved through the ring channel at very
high speed. From this remarkable inertia and resistance forces (damping forces) result.
The force effect of a hydro-mount is combined from the elasticity of the main spring and the
volume change in chamber 1.
With uF labelling the displacement of the generalized fluid mass MF ,

FH = cT xe + FF (xe uF ) (5.57)

holds, where the force effect of the main spring has been approximated by a linear spring with
the constant cT .
With MF R as actual mass in the ring channel and the cross sections AK , AR of chamber and
ring channel the generalized fluid mass is given by
 A 2
K
MF = MF R . (5.58)
AR

The fluid in chamber 1 is not being compressed, unless the membrane can evade no longer.
With the fluid stiffness cF and the membrane clearance sF one gets
 
c (xe uF ) + sF (xe uF ) < sF
F


FF (xe uF ) = 0 for |xe uf | sF (5.59)

 
c
F (xe uF ) sF (xe uf ) > +sF

The hard transition from clearance FF = 0 and fluid compression, resp. chamber deformation
with FF 6= 0 is not realistic and leads to problems, even with the numeric solution. The function
(5.59) is therefore smoothed by a parable in the range |xe uf | 2 sF .
The motions of the fluid mass cause friction losses in the ring channel, which are, at first
approximation, proportional to the speed,

FD = dF uF . (5.60)

The equation of motion for the fluid mass then reads as

MF uF = FF FD . (5.61)

The membrane clearing makes (5.61) nonlinear, and only solvable by numerical integration.
The nonlinearity also affects the overall force (5.57) in the hydro-mount.

88
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

Dynamic Stiffness [N/m] at Excitation Amplitudes A = 2.5/0.5/0.1 mm

400

300

200

100

Dissipation Angle [deg] at Excitation Amplitudes A = 2.5/0.5/0.1 mm

60

50

40

30

20

10

0
0 1
10 Excitation Frequency [Hz] 10

Figure 5.10: Dynamic Stiffness [N/mm] and Dissipation Angle [deg] for a Hydro-Mount

5.4.2.2 Dynamic Force Characteristics

The dynamic stiffness and the dissipation angle of a hydro bearing are displayed in Fig. 5.10
over the frequency. The dissipation angle is a measurement for the damping.
The simulation is based on the following system parameters

mF = 25 kg generalized fluid mass

cT = 125 000 N/m stiffness of main spring
dF = 750 N/(m/s) damping constant
cF = 100 000 N/m fluid stiffness
sF = 0.0002 mm clearance in membrane bearing

By the nonlinear and dynamic behavior a very good compromise between noise isolation and
vibration damping can be achieved.

89
Vehicle Dynamics FH Regensburg, University of Applied Sciences

5.5 Different Influences on Comfort and Safety

5.5.1 Vehicle Model

Ford motor company uses the vehicle dynamics program VeDynA (Vehicle Dynamic Analysis)
for comfort calculations.
The theoretical basics of the program modelling, generating the equations of motion, and
numeric solution have been published in the book G.Rill: Simulation von Kraftfahrzeugen,
Vieweg 1994
Through program extensions, adaption to different operating systems, installation of interfaces
to other programs and a menu-controlled in- and output, VeDynA has been subsequently de-
veloped to marketability by the company TESIS GmbH in Munich.
Ford
At the tire model tmeasy(tire model easy to use), as integrated in VeDynA, the tire forces are
calculated dynamically with respect to the tire deformation. For every tire a contact calculation
is made. The local inclination of the track is determined from three track points. From the
statistic characteristics of a track, spectral density and waviness, two-dimensional, irregular
tracks
Time =are calculated.
0.000000

ZZ

X
X Y
Y
Figure 5.11: Car Model

The vehicle model isExt.

Thilo Seibert specially
37598 distinguished by the following details:
/export/ford/dffa089/u/tseiber1/vedyna/work/results/mview.mvw 07/02/98
Vehicle Dynamics, Ford Research Center Aachen AA/FFA
nonlinear elastic kinematics of the wheel suspensions,
friction-affected and elastically suspended dampers,
fully elastic motor suspension by static and dynamic force elements
(rubber elements and/or hydro-mounts,
integrated passenger-seat models.
Beyond this, interfaces to external tire- and force element models are provided. A specially
developed integration procedure allows real-time simulation on a PC.

90
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

5.5.2 Simulation Results

The vehicle, a Ford Mondeo, occupied by two persons, drives with v = 80km/h over a country
road. The thereby occurring accelerations at the drivers seat rail and the wheel load variations
are displayed in Fig. 5.12.
The peak values of the accelerations and the maximal wheel load variations are arranged in
the tables 5.7 and 5.8 for the standard car and several modifications.

acceleration standard friction seat model engine mounts comfort bushings

xmin -0.7192 -0.7133 -0.7403 -0.5086 -0.7328
xmax +0.6543 +0.6100 +0.6695 +0.5092 +0.6886
ymin -1.4199 -1.2873 -1.4344 -0.7331 -1.5660
ymax +1.3991 +1.2529 +1.3247 +0.8721 +1.2564
zmin -4.1864 -3.9986 -4.1788 -3.6950 -4.2593
zmax +3.0623 2.7769 +3.1176 +2.8114 +3.1449

Table 5.7: Peak Acceleration Values

4Fz standard friction seat model engine mounts comfort bushings

front left 2.3830 2.4507 2.4124 2.3891 2.2394
front right 2.4208 2.3856 2.4436 2.3891 2.4148
rear left 2.1450 2.2616 2.1600 2.1113 2.1018
rear right 2.3355 2.2726 2.3730 2.2997 2.1608

Table 5.8: Wheel Load Variations 4Fz = Fzmax Fzmin

It can be seen, that the damper friction, the passengers, the engine suspension and the com-
pliance of the wheel suspensions, (here:represented by comfort bushings) influence especially
the accelerations and with this the driving comfort.
At fine tuning thus all these influences must be respected.

91
Vehicle Dynamics FH Regensburg, University of Applied Sciences

road profil [m] body longitudinal acceleration [m/s 2]

0.1 5

0.05

0 0

-0.05

-0.1 -5
0 500 1000 0 500 1000
[m] [m]

body vertical acceleration [m/s 2 ] body lateral acceleration [m/s 2 ]

5 5

0 0

-5 -5
0 500 1000 0 500 1000
[m] [m]

wheel load front left [kN] wheel load front right [kN]
6 6

5 5

4 4

3 3

2 2

1 1

0 0
0 500 1000 0 500 1000
[m] [m]
wheel load rear left [kN] wheel load rear right [kN]
6 6

5 5

4 4

3 3

2 2

1 1

0 0
0 500 1000 0 500 1000
[m] [m]

Figure 5.12: Road Profile, Accelerations and Wheel Loads

92
6 Driving Behavior of Single Vehicles

6.1 Standard Driving Maneuvers

6.1.1 Steady State Cornering

The steering tendency of a real vehicle is determined by the driving maneuver called steady
state cornering. The maneuver is performed quasi-static. The driver tries to keep the vehicle on
a circle with the given radius R. He slowly increases the driving speed v and, with this, because
2
of ay = vR , the lateral acceleration, until reaching the limit. Typical results are displayed in
Fig. 6.1.

80 4

60 2
side slip angle [deg]
steer angle [deg]

40 0

20 -2

0 -4
4 6

5
3
wheel loads [kN]
roll angle [deg]

2 3

2
1
1

0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
lateral acceleration [g] lateral acceleration [g]

Figure 6.1: Steady State Cornering: Rear-Wheel-Driven Car on R = 100 m

93
Vehicle Dynamics FH Regensburg, University of Applied Sciences

The vehicle is under-steering and thus stable. The inclination in the diagram steering angle
over lateral velocity decides, according to (4.122) with (4.123), about the steering tendency
and stability behavior.
The nonlinear influence of the wheel load on the tire performance is here used to design a ve-
hicle that is weakly stable, but sensitive to steer input in the lower range of lateral acceleration,
and is very stable but less sensitive to steer input in limit conditions.
With the increase of the lateral acceleration the roll angle becomes larger. The overturning
torque is intercepted by according wheel load differences between the outer and inner wheels.
With a sufficiently rigid frame the use of a anti roll bar at the front axle allows to increase the
wheel load difference there and to decrease it at the rear axle accordingly.
The digressive influence of the wheel load on the tire properties, cornering stiffness and max-
imally possible lateral force is thus stressed more strongly at the front axle and the vehicle
becomes more under-steering and stable at increasing lateral acceleration, until, in the limit
situation, it drifts out of the curve over the front axle.
Problems occur at front driven vehicles, because, due to the traction, the front axle cannot be
relieved at will.
Having a sufficiently 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.

6.1.2 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.
In Fig. 6.2 a real steering angle gradient is displayed.

40
steering angle [deg]

30

20

10

0
0 0.2 0.4 0.6 0.8 1
time [s]

Figure 6.2: Step Steer Input

Not the angle at the steering wheel is the decisive factor for the driving behavior, but the steer
angle at the wheels, which can differ from the steering wheel angle because of elasticities,

94
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

friction influences and a servo-support. At very fast steering movements also the dynamic
raise of 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.

0.6 12

0.5 10
lateral acceleration [g]

0.4 8

yaw velocity [deg/s]

0.3 6

0.2 4

0.1 2

0 0
3 1

2.5 0.5
side slip angle [deg]

2 0
roll angle [deg]

1.5 -0.5

1 -1

0.5 -1.5

0 -2
0 2 4 0 2 [t] 4

Figure 6.3: Step Steer: Passenger Car at v = 100 km/h

The vehicle behaves dynamically very well, Fig. 6.3. Almost no exceeds at roll angle and lateral
acceleration. Small exceeds at yaw velocity and sideslip angle.

6.1.3 Driving Straight Ahead

6.1.3.1 Random Road Profile

The irregularities of a track are of stochastic nature. Fig. 6.4 shows a country road profile in
different scalings. To limit the effort at 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 ) (6.1)

95
Vehicle Dynamics FH Regensburg, University of Applied Sciences

0.05
0.04
0.03
0.02
0.01
0
-0.01
-0.02
-0.03
-0.04 5
4
-0.05 3
2
0 1
10 20 30 40 50 60 70 80 90 100 0

Figure 6.4: Track Irregularities

or one uses an isotropic track. At an isotropic track the statistic properties are direction-
independent. Then a two-dimensional track with its stochastic properties can be described
by a single random process
z = z(x, y) z = z(s) ; (6.2)

A normally distributed, stationary and ergodic random process z = z(s) is completely charac-
terized by the first two expectation values, mean value
Zs
1
mz = lim z(s) ds (6.3)
s 2s
s

and correlating function

Zs
1
Rzz () = lim z(s) z(s ) ds (6.4)
s 2s
s

. A vanishing mean value mz = 0 can always be achieved by an appropriate coordinate

transformation. The correlation function is symmetric,
Rzz () = Rzz () (6.5)
and
Zs
1
2
Rzz (0) = lim z(s) ds (6.6)
s 2s
s
describes the squared average 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 (6.7)
0

96
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

where denotes the space circular frequency. With (6.7) follows from (6.6)

Z
Rzz (0) = Szz () d . (6.8)
0

The psd thus gives information, how the square average is compiled from the single frequency
shares.
The power spectral densities of real tracks can be approximated by the relation1
 w

Szz () = S0 (6.9)
0

Where the reference frequency is fixed to 0 = 1 m1 . The reference psd S0 = Szz (0 ) acts
as a measurement for unevennes and the waviness w indicates, whether the track has notable
irregularities in the short or long wave spectrum.
At real tracks reference-psd and waviness lie within the range

1 106 m3 S0 100 106 m3 and

6.1.3.2 Steering Activity

A straightforward drive upon an uneven track makes continuous steering corrections neces-
sary. The histograms of the steering angle at a driving speed of v = 90 km/h are displayed in
Fig. 6.5.

-6 3 -5 3
highway: S 0=1*10 m ; w=2 country road: S0=2*10 m ; w=2
1000 1000

500 500

0 0
-2 0 [deg] 2 -2 0 [deg] 2

Figure 6.5: Steering Activity on different Roads

The track quality is reflected in the amount of steering actions. The steering activity is often
used to judge a vehicle in practice.
1
cf.: M. Mitschke: Dynamik der Kraftfahrzeuge (Band B), Springer-Verlag, Berlin 1984, S. 29.

97
Vehicle Dynamics FH Regensburg, University of Applied Sciences

6.2 Coach with different Loading Conditions

6.2.1 Data

At trucks and coaches the difference between empty and laden is sometimes very large. In the
table 6.1 all relevant data of a travel coach in fully laden and empty condition are arrayed.

vehicle mass [kg] center of gravity [m] inertias [kg m2 ]

12 500 0 0
empty 12 500 3.800 | 0.000 | 1.500 0 155 000 0
0 0 155 000
15 400 0 250
fully laden 18 000 3.860 | 0.000 | 1.600 0 200 550 0
250 0 202 160
Table 6.1: Data for a Laden and Empty Coach

The coach has a wheel base of a = 6.25 m. The front axle with the track width sv = 2.046 m
has a double wishbone single wheel suspension. The twin-tire rear axle with the track widths
soh = 2.152 m and sih = 1.492 m is guided by two longitudinal links and an a-arm. The air-
springs are fitted to load variations via a niveau-control.

6.2.2 Roll Steer Behavior

10
suspension travel [cm]

-5

-10
-1 0 1
steer angle [deg]

Figure 6.6: 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 steer effect, Fig. 6.6.

98
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

6.2.3 Steady State Cornering

Fig. 6.7 shows the results of a steady state cornering on a 100 m-Radius.
The fully occupied vehicle is slightly more understeering than the empty one. The higher wheel
loads cause greater tire aligning torques and increase the digressive wheel load influence on
the increase of the lateral forces. Additionally roll steering at the rear axle occurs.

steer angle [deg] vehicle course

LW
250 200

200 150

[m]
150 100

100 50

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

wheel loads [kN] wheel loads [kN]

100 100

50 50

0 0
0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4
lateral acceleration a y [g] lateral acceleration a y [g]

Figure 6.7: Steady State Cornering: Coach - - empty, fully occupied

In the limit range both vehicles can not be kept on the given radius. Due to the high position
of the center of gravity the maximal lateral acceleration is limited by the overturning hazard.
At the empty vehicle, the inner front wheel lift off at a lateral acceleration of ay 0.4 g . If the
vehicle is fully occupied, this effect occurs already at ay 0.35 g .

6.2.4 Step Steer Input

The results of a step steer input at the driving speed of v = 80 km/h can be seen in Fig. 6.8.
To achieve comparable acceleration values in steady state condition, the step steer input was
done at the empty vehicle with = 90 Grad and at the fully occupied one with = 135 Grad.
The steady state roll angle is at the fully occupied bus 50% larger than at the empty one.
By the niveau-control the air spring stiffness increases with the load. Because the damper
effect remains unchange, the fully laden vehicle is not damped as well as the empty one. The
results are higher exceeds in the lateral acceleration, the yaw speed and sideslip angle.

99
Vehicle Dynamics FH Regensburg, University of Applied Sciences

lateral acceleration a y [g] yaw velocity Z [deg/s]

0.4 10

8
0.3
6
0.2
4
0.1
2

0 0
0 2 4 6 8 0 2 4 6 8

roll angle [deg] side slip angle [deg]

8
2
6
1

4 0

-1
2
-2
0
0 2 4 [s] 6 8 0 2 4 [s] 6 8

Figure 6.8: Step Steer: - - Coach empty, Coach fully occupied

6.3 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.
The semi-trailing axle realized in serial production represents, according to the roll support,
Fig. 6.9, a compromise between the trailing arm and the single wishbone.

10
vertical motion [cm]

-5

-10
-5 0 5
lateral motion [cm]

Figure 6.9: Rear Axle Kinematics: Semi-Trailing Arm, - - Single Wishbone, Trailing Arm

The influences on the driving behavior at steady state cornering on a 100 m radius are shown
in Fig. 6.10.

100
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

steer angle [deg] roll angle [Grad]

LW
100 5
4
3
50
2
1
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8

wheel loads front [kN] wheel loads rear [kN]

6 6

4 4

2 2

0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
lateral acceleration a y [g] lateral acceleration a y [g]

Figure 6.10: Steady State Cornering, Semi-Trailing Arm, - - Single Wishbone, Trailing
Arm

Substituting the semi-trailing arm at the standard car by a single wishbone, one gets, without
adaption of the other system parameters, a vehicle, which oversteers in the limit range.
The single wishbone causes, compared to the semi-trailing arm a notably higher roll support.
This increases the wheel load difference at the rear axle, Fig. 6.10. Because the wheel load
difference is simultaneously reduced at the front axle, the understeer tendency is reduced. In
the limit range, this even leads to oversteer behavior.
The vehicle with a trailing arm rear axle is, compared to the serial car, more understeering.
The lack of roll support at the rear axle also causes a larger roll angle.

101
Index

Ackermann Geometry, 45 Driving Safety, 73

Ackermann Steering Angle, 45, 69 Dynamic Axle Load, 31
Aerodynamic Forces, 32 Dynamic Force Elements, 84
Air Resistance, 32 Dynamic Wheel Loads, 30
All Wheel Drive, 61
Angular Wheel Velocity, 27 Eigenvalues, 65, 75
Anti Dive, 44 Environment, 3
Anti Roll Bar, 57
Anti Squat, 44 First Harmonic Oscillation, 84
Anti-Lock-Systems, 38 FourierApproximation, 85
Axle Kinematics, 44 Free Vibrations, 76
Double Wishbone, 10 Frequency Domain, 84
McPherson, 10 Friction, 33
Multi-Link, 10 Front Wheel Drive, 34, 61
Axle Load, 31
Generalized Fluid Mass, 88
Axle Suspension
Grade, 31
Rigid Axle, 5
Twist Beam, 5 Hydro-Mount, 87
Bend Angle, 52
Kingpin, 10
Brake Pitch Angle, 39
Kingpin Angle, 11
Brake Pitch Pole, 44
Kingpin Inclination, 11
Kingpin Offset, 12
Camber Angle, 9, 17
Camber Compensation, 56, 58
Lateral Acceleration, 55, 69
Camber Slip, 26
Lateral Force, 62
Caster Angle, 11
Lateral Slip, 62, 63
Caster Offset, 12
Load, 3
Characteristic Speed, 68
Climbing Capacity, 33 Maximum Acceleration, 33, 34
Comfort, 73 Maximum Deceleration, 33, 35
Contact Geometry, 16
Contact Point, 18 Optimal Brake Force Distribution, 36
Contact Point Velocity, 19 Optimal Damper, 81
Cornering Resistance, 60, 61 Optimal Damping, 76, 78
Cornering Stiffness, 23, 69 Optimal Drive Force Distribution, 36
Curve Radius, 45 Optimal Parameter, 81
Optimal Spring, 81
Damper Characteristic, 79 Oversteer, 69
Disturbing Force Lever, 12 Overturning Limit, 53
Down Forces, 32
Drag Link, 6, 7 Parallel Tracks, 95
Drive Pitch Angle, 39 Pinion, 6
Driver, 2 Power Spectral Density, 96
Driving Preload, 74
Maximum Acceleration, 34
Driving Comfort, 77 Quality Criteria, 80

i
Vehicle Dynamics FH Regensburg, University of Applied Sciences

Quarter Car Model, 78 Dynamic Offset, 24

Dynamic Radius, 15, 16
Rack, 6 Lateral Direction, 17
Random Road Profile, 79, 95 Lateral Force, 14
Rear Wheel Drive, 34, 61 Lateral Force Distribution, 23
Referencies Lateral Slip, 23
Hirschberg, W., 28 Lateral Velocity, 19
Rill, G., 28 Linear Model, 62
Weinfurter, H., 28 Loaded Radius, 16, 17
Road, 17 Longitudinal Force, 14, 21, 22
Roll Axis, 58 Longitudinal Force Characteristics, 22
Roll Center, 58 Longitudinal Force Distribution, 22
Roll Steer, 98 Longitudinal Slip, 22
Roll Stiffness, 55 Longitudinal Velocity, 19
Roll Support, 56, 58 Normal Force, 14
Rolling Condition, 62 Pneumatic Trail, 24
Radial Damping, 20
Safety, 73 Radial Direction, 17
Side Slip Angle, 45 Radial Stiffness, 55
Space Requirement, 46 Rolling Resistance, 14
Spring Characteristic, 79 Self Aligning Torque, 14, 24
Spring Rate, 75 Sliding Velocity, 23
Stability, 65 Static Radius, 16, 17
State Equation, 65 Tilting Torque, 14
Steady State Cornering, 53, 93, 99 Transport Velocity, 16
Steer Box, 6, 7 Tread Deflection, 21
Steer Lever, 7 Tread Particles, 21
Steering Activity, 97 Undeformed Radius, 16
Steering Angle, 51 Vertical Force, 20
Steering System tire composites, 13
Drag Link Steering, 7 Tire Development, 13
Lever Arm, 6 Tire Model
Rack and Pinion, 6 Kinematic, 45
Steering Tendency, 59, 69 Linear, 70
Step Steer Input, 94, 99 TMeasy, 28
Suspension Model, 73 Tire Stiffness
Suspension Spring Rate, 75 Radial, 20
Sweep-Sine, 86 Toe Angle, 9
System Response, 84 Track, 74
Track Curvature, 51
Tilting Condition, 33 Track Normal, 17, 18
Tire Track Radius, 51
Bore Slip, 28 Track Width, 45, 55
Bore Torque, 14, 27, 28 Trailer, 48, 51
Camber Angle, 17 Turning Center, 45
Camber Influence, 25
Characteristics, 28 Understeer, 69
Circumferential Direction, 17
Contact Area, 14 Vehicle, 2
Contact Forces, 14 Vehicle Comfort, 73
Contact Length, 22 Vehicle Data, 80
Contact Point, 16 Vehicle Dynamics, 1
Contact Torques, 14 Vehicle Model, 30, 39, 48, 56, 62, 73, 78, 90
Cornering Stiffness, 24 Virtual Work, 57
Deflection, 17, 22
Deformation Velocity, 19 Waviness, 97

ii
FH Regensburg, University of Applied Sciences Prof. Dr.-Ing. G. Rill

Wheel Base, 45
Wheel Load, 14
Wheel Loads, 30
Wheel Suspension
Central Control Arm, 5
Double Wishbone, 4
McPherson, 4
Multi-Link, 4
Semi-Trailing Arm, 5, 100
Single Wishbone, 100
SLA, 5
Trailing Arm, 100

Yaw Angle, 51
Yaw Velocity, 63

iii