Professional Documents
Culture Documents
Vehicle Dynamics - Rill PDF
Vehicle Dynamics - Rill PDF
Vehicle Dynamics - Rill PDF
FACHHOCHSCHULE REGENSBURG
UNIVERSITY OF APPLIED SCIENCES
HOCHSCHULE FR
TECHNIK
WIRTSCHAFT
SOZIALES
LECTURE NOTES
Prof. Dr. Georg Rill
October 2005
download: http://homepages.fh-regensburg.de/%7Erig39165/
Contents
Contents
1
Introduction
1.1
Literature
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2
Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1
Vehicle Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2
Driver
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.3
Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.4
Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.5
Environment
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3
Denitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.1
Reference frames
1.3.2
1.3.3
1.3.4
Steering Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.4.1
Kingpin . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.4.2
1.3.4.3
. . . .
Road
10
2.1
Modeling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
2.2
2.3
Deterministic Proles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.2.1
. . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.2.2
Sine Waves
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
2.3.1
Random Proles
Statistical Properties . . . . . . . . . . . . . . . . . . . . . . . . . .
12
2.3.2
15
2.3.3
Realizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16
2.3.3.1
Sinusoidal Approximation
. . . . . . . . . . . . . . . . . .
16
2.3.3.2
Shaping Filter
. . . . . . . . . . . . . . . . . . . . . . . .
17
2.3.3.3
Two-Dimensional Model . . . . . . . . . . . . . . . . . . .
18
Tire
3.1
3.2
3.3
3.4
19
3.1.1
Tire Development . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
3.1.2
Tire Composites
19
3.1.3
. . . . . . . . . . . . . . . . . . . . . . . .
20
3.1.4
21
Contact Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
3.2.1
Basic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
3.2.2
25
3.2.3
Tire Deection
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
26
3.2.4
28
3.2.5
. . . . . . . . . . . . . . . . . . . . . . . . . .
29
3.2.6
31
3.2.7
. . . . . . . . . . . . . . . . . . . . . . . .
32
34
3.3.1
Wheel Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
34
3.3.2
Tipping Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35
3.3.3
Rolling Resistance
. . . . . . . . . . . . . . . . . . . . . . . . . . .
36
. . . . . . . . . . . . . . . . . . . . . . . . . .
37
3.4.1
37
3.4.2
40
3.4.3
41
3.4.4
43
3.4.5
. . . . . . . . . . . . . . . . . . . . .
45
3.4.6
. . . . . . . . . . . . . . . . . . . . . . . . . .
46
3.4.7
Camber Inuence . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
3.4.8
Bore Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
3.4.9
52
. . . . . . . . . . . . . . . . . . . . . .
Suspension System
54
4.1
. . . . . . . . . . . . . . . . . . . . . . . . . . .
54
4.2
Some Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
4.2.1
55
4.2.2
Specic Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
56
4.3
4.4
II
19
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Steering Systems
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
56
4.3.1
Requirements
4.3.2
56
4.3.3
4.3.4
4.3.5
58
. . . . . . . . . . . . . . . . . . . . . . .
57
. . . . . . . . . . . . . . . . . . . . . .
57
58
59
4.4.1
Springs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
59
4.4.2
Damper
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
60
4.4.3
Rubber Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . .
62
4.5
63
. . . . . . . . . . . . . . . . . .
63
4.5.2
. . . . . . . . . . . . . . . . .
66
4.5.3
68
4.5.3.1
69
Hydro-Mount . . . . . . . . . . . . . . . . . . . . . . . . .
Vertical Dynamics
72
5.1
Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
72
5.2
Modelling Aspects
72
5.3
5.4
5.5
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.1
5.2.2
Twodimensional Models
. . . . . . . . . . . . . . . . . . . . . . . .
73
5.2.3
Simple Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
75
Basic Tuning
. . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3.1
5.3.2
Spring Rates
6.2
6.3
6.4
76
. . . . . . . . . . . . . . . .
76
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
78
5.3.2.1
78
5.3.2.2
Nonlinear Springs . . . . . . . . . . . . . . . . . . . . . . .
79
5.3.3
Inuence of Damping . . . . . . . . . . . . . . . . . . . . . . . . . .
81
5.3.4
Optimal Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . .
81
5.3.4.1
Avoiding Overshoots . . . . . . . . . . . . . . . . . . . . .
81
5.3.4.2
. . . . . . . . . . . . . . .
82
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
87
Modelling Aspects
. . . . . . . . . . . . . . . . . . . . . . . . . . .
87
5.4.2
88
5.4.3
Technical Realization . . . . . . . . . . . . . . . . . . . . . . . . . .
90
. . . . . . . . . . . . . . . . . . . . . . . . . . .
90
5.5.1
90
5.5.2
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
92
Longitudinal Dynamics
6.1
72
94
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
94
6.1.1
94
6.1.2
Inuence of Grade
. . . . . . . . . . . . . . . . . . . . . . . . . . .
95
6.1.3
Aerodynamic Forces
. . . . . . . . . . . . . . . . . . . . . . . . . .
96
Maximum Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
6.2.1
Tilting Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
6.2.2
Friction Limits
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
97
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
98
6.3.1
98
6.3.2
99
6.3.3
6.3.4
6.3.5
Anti-Lock-Systems
. . . . . . . . . . . . . . . . 102
. . . . . . . . . . . . . . . . . . . . . . . . . . . 102
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
III
6.4.1
6.4.2
Equations of Motion
6.4.3
Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.4.4
6.4.5
7.2
7.3
109
Kinematic Approach
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.1
. . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.2
Ackermann Geometry
. . . . . . . . . . . . . . . . . . . . . . . . . 109
7.1.3
7.1.4
Kinematics
7.1.4.2
Vehicle Motion
. . . . . . . . . . . . . . . . . . . . . . . . . . 112
7.1.4.3
7.1.4.4
7.1.4.5
Course Calculations
. . . . . . . . . . . . . . . . . . . . . 116
7.2.2
Overturning Limit
7.2.3
7.2.4
7.2.5
Wheel Loads
. . . . . . . . . . . . . . . . . . . . . . . . . . . 118
. . . . . . . . . . . . . . . . . . . . . . . 124
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Modeling Concept
7.3.2
Kinematics
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.3
Tire Forces
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.3.4
7.3.5
Equations of Motion
7.3.6
Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.3.8
. . . . . . . . . . . . . . . . . . . . . . . . . . . 125
. . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.3.6.1
Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . 128
7.3.6.2
7.3.6.3
. . . . . . . . . . . . . . . . . 129
7.3.7.2
Steering Tendency
7.3.7.3
Slip Angles
. . . . . . . . . . . . . . . . . . . . . . 132
. . . . . . . . . . . . . . . . . . . . . . . . . . 133
. . . . . . . . . . . . . . . . . . . . . . . . 113
7.2.1
7.3.7
IV
. . . . . . . . . . . . . . . . . . . . . . . . . . 107
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Lateral Dynamics
7.1
. . . . . . . . . . . . . . . . . . . . . . . . . . 105
. . . . . . . . . . . 133
136
8.1.2
8.1.3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
8.2
8.3
8.1.3.1
8.1.3.2
. . . . . . . . . . . . . . . . . . . . 138
. . . . . . . . . . . . . . . . . . . 140
8.2.1
Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
8.2.2
Roll Steering
8.2.3
8.2.4
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
. . . . . . . . . . . . . . 143
1 Introduction
1.1 Literature
M.;
Wallentowitz,
Blundell, M.; Harty, D.: The Multibody System Approach to Vehicle Dynamics.
Kraftfahrtechnisches
Handbuch:
Robert
Bosch
GmbH
(Hrsg.),
23.
Au.,
Vieweg-Verlag.
Proceedings:
Vehicle Dynamics
1.2 Terminology
1.2.1 Vehicle Dynamics
The expression `Vehicle Dynamics' encompasses the interaction of:
driver
vehicle
load
environment
computer calculations
eld tests
In the following the interactions between the single systems and the problems with computer calculations and/or measurements shall be discussed.
1.2.2 Driver
By various means the driver can interfere with the vehicle:
driver
steering wheel
accelerator pedal
brake pedal
clutch
gear shift
lateral dynamics
longitudinal dynamics
vehicle
vehicle
vibrations:
sounds:
instruments:
driver
environment
climate
track
trac density
driver
The driver's reaction is very complex. To achieve objective results, an `ideal' driver is
used in computer simulations, and in driving experiments automated drivers (e.g. steering
machines) are employed.
Transferring results to normal drivers is often dicult, if eld tests are made with test
drivers. Field tests with normal drivers have to be evaluated statistically. Of course, the
driver's security must have absolute priority in all tests.
Driving simulators provide an excellent means of analyzing the behavior of drivers even
in limit situations without danger.
It has been tried to analyze the interaction between driver and vehicle with complex driver
models for some years.
1.2.3 Vehicle
The following vehicles are listed in the ISO 3833 directive:
motorcycles
passenger cars
busses
trucks
agricultural tractors
road trains
Vehicle Dynamics
1.2.4 Load
Trucks are conceived for taking up load. Thus, their driving behavior changes.
Load
In computer calculations problems occur at the determination of the inertias and the
modeling of liquid loads.
Even the loading and unloading process of experimental vehicles takes some eort. When
carrying out experiments with tank trucks, ammable liquids have to be substituted with
water. Thus, the results achieved cannot be simply transferred to real loads.
1.2.5 Environment
The environment inuences primarily the vehicle:
Environment
road:
air:
vehicle
environment
climate
visibility
driver
Through the interactions between vehicle and road, roads can quickly be destroyed.
The greatest diculty with eld tests and laboratory experiments is the virtual impossibility of reproducing environmental inuences.
The main problems with computer simulation are the description of random road irregularities and the interaction of tires and road as well as the calculation of aerodynamic
forces and torques.
1.3 Definitions
1.3.1 Reference frames
A reference frame xed to the vehicle and a ground-xed reference frame are used to
describe the overall motions of the vehicle, Figure 1.1. The ground-xed reference frame
The wheel rotates around an axis which is xed to the wheel carrier. The reference frame
F.
xC , yC
and
zC
are parallel to
zF
yF
xF
z0
zC
x0
yC
en
xC
y0
eyR
The momentary position of the wheel is xed by the wheel center and the orientation of
the wheel rim center plane which is dened by the unit vector
eyR
en
xF
left
wheel
yF
vehicle
center
plane
right
wheel
rear
Figure 1.2: Positive toe-in angle
According to the DIN 70 000 directive the angle
longitudinal direction and the intersection line of the tire center plane with the track
plane is named toe or toe-in angle. It will be positive, if the front part of the wheel is
oriented towards the vehicle center plane, Figure 1.2. Toe-in reduces the tendency of the
wheels to shimmy.
The camber angle
en .
is the angle between the wheel center plane and the local track normal
It will be positive, if the upper part of the wheel is inclined outwards, Figure 1.3. A
Vehicle Dynamics
en
en
left
wheel
right
wheel
zF
yF
road
eyR
describes the wheel rotation axis. Its orientation with respect to the
wheel carrier xed reference frame can be dened by the angles 0 and 0 or 0 and 0 ,
Fig. 1.4. In design position the corresponding axes of the frames
and
are parallel.
zC = zF
xC = xF
0
0
0*
yC = yF
eyR
eyR,F = eyR,C
tan 0
1
1
= q
tan2 0 + 1 + tan2 0
tan 0
or
eyR,F = eyR,C
(1.2)
where
sin 0 cos 0
= cos 0 cos 0 ,
sin 0
(1.1)
xF - yF -plane,
the angle
en
camber angle 0 . To specify the dierence between 0 and 0 the ratio between the third
the vertical axes
the angles
0 .
0 .
zC = zF
and
eyR
sin 0
tan 0
=
1
cos 0 cos 0
0 1
or
tan 0 =
tan 0
.
cos 0
(1.3)
and
is hardly
noticeable.
zR
B
yR
M
xR
and
kingpin is no longer dened by real joints. Here, as well as with the enhanced McPherson
wheel suspension, the kingpin changes its position relative to the wheel body at wheel
travel and steering motions.
Vehicle Dynamics
B
zR
zR
yR
yR
xR
xR
M
A
rotation axis
eS describes the direction of the kingpin axis. Within the vehicle xed
reference frame F it can be xed by two angles. The caster angle denotes the angle
between the zF -axis and the projection line of eS into the xF -, zF -plane. In a similar
way the projection of eS into the yF -, zF -plane delivers the kingpin inclination angle ,
Fig. 1.7.
eS
zF
zF
yF
xF
eS
in the
(1)
tan =
eS,F
(3)
eS,F
where
(1)
(2)
(3)
(2)
and
tan =
eS,F
F.
(1.4)
eS,F
(3)
eS,F
P,
en
ex
and
ey
which
point into the direction of the longitudinal and lateral tire force result from the contact
geometry. The axle kinematics denes the kingpin line. In general, the point
where an
extension oft the kingpin line meets the road surface does not coincide with the contact
point
P,
Fig. 1.8. As both points are located on the local track plane, for the left wheel
to
can be written as
rSP = c ex + s ey ,
where
positive, if
(1.5)
Caster and steering oset will be
P.
kingpin line
C d
ey
en
P
ex
S
2 Road
2.1 Modeling Aspects
Sophisticated road models provide the road height
at each point
x, y ,
zR
Fig. 2.1.
Road profile
z0
y0
Segments
z(x,y)
x0
(x,y)
Friction
Obstacle
s = s(t).
10
z
x
z2
z1
zR(x,y)
z1(s)
Now, the road heights on the left and right track are provided by two one-dimensional
functions
z1 = z1 (s)
z2 = z2 (s).
and
about the local lateral road inclination is available. If this information is not provided by
additional functions the impact of a local lateral road inclination to vehicle motions is not
taken into account.
For basic studies the irregularities at the left and the right track can considered to be
z1 (s) z2 (s).
z1 (x) = z2 (x)
zR (s) =
can be used. Now, the roll excitation of the vehicle is neglected too.
H
B
y
L
(
z(x, y) =
H
0
if
else
0<x<L
and
12 B < y < 12 B
(2.1)
11
Vehicle Dynamics
z(x, y) =
1
2
where
H, B
if negative
x
H 1 cos 2
L
0
0<x<L
if
and
21 B < y < 12 B
(2.2)
else
L denote height, width and length of the obstacle. Potholes are obtained
values for the height (H < 0) are used.
and
z2 (s) = A sin ( s ) ,
z1 (s) = A sin ( s) ,
(2.3)
s is the path variable, A denotes the amplitude, the wave number, and the angle
describes a phase lag between the left and the right track. The special cases = 0 and
= represent the in-phase excitation with z1 = z2 and the out of phase excitation with
z1 = z2 .
where
s = v0 t,
where the
(2.4)
2
2
v0
s=
v0 t = 2 t = t .
L
L
L
(2.5)
L =
the term
can be written as
s =
f = /(2) = v0 /L.
0.5 Hz to 15 Hz . This
v0 /L 0.5 Hz and v0 /L 15 Hz .
For most of the vehicles the rigid body vibrations are in between
range is covered by waves which satisfy the conditions
whole frequency range with moderate vehicle velocities proles with dierent varying
wavelengths are needed.
12
zR
0.15
Gaussian
density
function
Realization
0.10
[m]
0.05
0
-0.05
-0.10
-0.15
-200
Histogram
s
-100
-150
-50
50
200
[m] 150
100
1
m = E {zR (s)} = lim
X X
ZX/2
zR (s) ds = 0
(2.6)
X/2
can be achieved, where
p(zR ) =
zR2
e 2 2 ,
(2.7)
zR = zR (s)
ZX/2
1
2 = E zR2 (s) = lim
X X
zR (s)2 ds .
(2.8)
X/2
Alteration of
inexion occur at
P () =
|z| <
is given by
Z+ z 2
e 2 2 dz .
(2.9)
and
P (3) = 0.997.
1
R() = E {zR (s) zR (s+)} = lim
X X
ZX/2
zR (s) zR (s+) ds .
(2.10)
X/2
13
Vehicle Dynamics
R() = R(),
zR (s) and zR (s+) diminishes. For large values of the two values are practically
unrelated. Hence, R( ) will tend to 0. In fact, R() is always less R(0), which
2
coincides with the variance of the process. If a periodic term is present in the process
it will show up in R().
between
Usually, road proles are characterized in the frequency domain. Here, the auto-correlation
function
S()
R()
S().
In general,
R()
and
1
S() =
2
R() e
1
R() =
2
and
where
S() ei d ,
(2.11)
i is the imaginary unit, and in rad/m denotes the wave number. To avoid negative
() = 2 S() ,
the relationship
if
() = 0 ,
and
ei = cos() i sin(),
if
<0,
(2.12)
R() = R()
2
() =
Z
R() cos () d
and
Z
R() = () cos () d .
(2.13)
Z
= R( = 0) = () d .
2
(2.14)
0
In reality the psd
()
1 N ,
(i)
N
X
i=1
14
(i ) 4
with
4 =
N 1
.
N
(2.15)
() = (0 )
where,
= 2/L
in
rad/m
w
(2.16)
0 = (0 ) in m2 /(rad/m)
number 0 = 1 rad/m. The drop
10-3
w.
10-4
10-5
Class E
0=1106
10-6
10-7
10-8
10-9 -2
10
Class A
10-1
100
101
Wave number [rad/m]
102
10-2
10-1
100
101
Wave number [rad/m]
102
15
Vehicle Dynamics
2.3.3 Realizations
2.3.3.1 Sinusoidal Approximation
A random prole of a single track can be approximated by a superposition of
sine waves
zR (s) =
N
X
Ai sin (i s i ) ,
(2.17)
i=1
Ai and its
i , i = 1(1)N
i .
wave number
and
By
dierent proles can be generated which are similar in the general appearance
1
2 = lim
X X
ZX/2 X
N
X/2
For
i=j
and for
i 6= j
!
Ai sin (i s i )
i=1
N
X
!
Aj sin (j s j )
ds .
(2.18)
j=1
i=j
can
be solved immediately
Z
Jii =
A2i
1
A2i
i si sin 2 (i si ) .
sin (i si ) ds =
2i
2
2
(2.19)
sin x sin y =
the integrals for
i 6= j
1
1
cos(xy) cos(x+y)
2
2
(2.20)
Z
Jij =
Ai sin (i si ) Aj sin (j sj ) ds
1
= A i Aj
2
=
1
cos (ij s ij ) ds Ai Aj
2
Z
cos (i+j s i+j ) ds
(2.21)
1 Ai Aj
1 A i Aj
sin (ij s ij ) +
sin (i+j s i+j )
2 ij
2 i+j
ij = ij
and
ij = ij
1.
results in
N
N
N
1 X X/2
1 X X/2
1X 2
Jii X/2 + lim
Jij X/2 =
= lim
A .
X X
X X
2 i=1 i
i=1
i,j=1
|
{z
}
|
{z
}
N
2
0
X A
i
i
2
i
i=1
2
16
(2.22)
On the other hand, the variance of a sinusoidal approximation to a random road prole
is given by Eq. (2.15). So, a road prole
given psd
()
Ai =
and the wave numbers
2 (i ) 4 ,
i = 1(1)N ,
equal intervals
(2.23)
4.
0.10
[m]
zR = zR (s)
0.05
0
-0.05
-0.10
0
10
20
30
40
50
60
70
80 [m] 90
100
where
is a constant, and
d
zR (s) = zR (s) + w(s) ,
ds
w(s) is a white noise process with
zR
reads as
(2.24)
w .
R = H() W H T () =
where
H()
1
W
1
W
= 2
,
+ i
i
+ 2
(2.25)
lter.
By setting
17
Vehicle Dynamics
10-3
Measurements
Shaping filter
10-4
10-5
10-6
10-7
10-8
10-9 -2
10
10-1
100
101
Wave number [rad/m]
102
zR = zR (x, y)
proximation is very laborious. Because a shaping lter is a dynamic system, the resulting road prole realizations are not reproducible. By adding band-limited white noise
processes and taking the momentary position
x, y
z
x
1
0
-1
4
0
-2
-4
10
15
20
25
30
35
40
45
50
18
3 Tire
3.1 Introduction
3.1.1 Tire Development
Some important mile stones in the development of tires are shown in Table 3.1.
1839
1845
1888
1893
1895
50 tire deations,
22 complete inner tube changes
1899
1904
1908
1922
1943
1946
Radial Tire
Table 3.1: Milestones in tire development
Of course the tire development did not stop in 1946, but modern tires are still based on
this achievements.
8.5 kg
19
Vehicle Dynamics
16%
Rubber: natural/synthetic
38%
30%
10%
4%
Miscellaneous
2%
www.felge.de
140 mm
180 mm
2.5
bar,
Fz = 4700 N
The eect of the contact forces can be fully described by a resulting force vector applied
at a specic point of the contact patch and a torque vector. The vectors are described in a
track-xed reference frame. The
to the
z -axis
z -axis
eyR .
x-axis
is perpendicular
y -axis.
The components of the contact force vector are named according to the direction of the
axes, Figure 3.2.
A non symmetric distribution of the forces in the contact patch causes torques around
the
and
Tx .
The torque
the rolling resistance of the tire. In particular, the torque around the
z -axis
Ty
includes
is important
Tz = TB + TS .
20
(3.1)
Fx
Fy
Fz
longitudinal force
Tx
Ty
Tz
tilting torque
lateral force
vertical force or wheel load
eyR
Fy
Tx
Fx
Tz
Ty
Fz
MS
z -axis
MB .
takes into account that ,in general, the resulting lateral force is not acting in
Test trailer
tire
exact contact
real road
compensation wheel
test wheel
21
Vehicle Dynamics
tire
too small
contact area
safety walk
coating
tire
perfect contact
rotation
drum
tire
On drum test rigs the tire is placed either inside or outside of the drum. In both cases
the shape of the contact area between tire and drum is not correct. That is why, one
can not rely on the measured self aligning torque. Due its simple and robust design, wide
applications including measurements of truck tires are possible.
The at bed tire test rig is more sophisticated. Here, the contact patch is as at as on the
road. But, the safety walk coating which is attached to the steel bed does not generate
the same friction conditions as on a real road surface.
Radial 205/50 R15, FN= 3500 N, dry asphalt
4000
3000
2000
Driving
1000
0
-1000
Braking
-2000
-3000
-4000
-40
-30
-20
-10
10
20
30
40
Tire forces and torques are measured in quasi-static operating conditions. Hence, the measurements for increasing and decreasing the sliding conditions usually result in dierent
graphs, Figure 3.5. In general, the mean values are taken as steady state results.
22
eyR
is given by the
3.6.
rim
centre
plane e zR
tire
e yR
rMP
e yR
en
wheel
carrier
ex
b
P0 a
ey P
en
P0
z0
P*
x0
road: z = z ( x , y )
x0
y0
y0
z0
The irregularities of the track can be described by an arbitrary function of two spatial
coordinates
z = z(x, y).
At an uneven track the contact point
(3.2)
rM P = r0 ezB ,
where
r0
ezB
(3.3)
z -direction of the
z0
x0 , y0 ,
is determined by
x
= y ,
z
r0P ,0 = r0M,0 + rM P ,0
(3.4)
23
Vehicle Dynamics
r0M
P0
r0P0 ,0 =
M.
follows from
(3.5)
z (x , y )
where Eq. (3.2) was used to calculate the appropriate road height. In the point
track normal
en
P0
the
ex =
where
ex
eyR en
| eyR en |
ey = en ex ,
and
(3.6)
eyR denotes the unit vector into the direction of the wheel rotation axis. Calculating
eyR not always being perpendicular to the track. The tire
demands a normalization, as
camber angle
= arcsin eTyR en
(3.7)
describes the inclination of the wheel rotation axis against the track normal.
The vector from the rim center
P0
rM P0 = rS ezR + a ex + b ey ,
where
rS
a, b
(3.8)
ferential and lateral direction, and the radial direction is given by the unit vector
ezR = ex eyR
which is perpendicular to
ex
and
eyR .
(3.9)
en
results
in
ex
and
ey
are perpendicular to
en
(3.10)
(3.11)
rS
The contact point
eTn rM P0
= T
.
en ezR
(3.12)
rM P = rS ezR
(3.13)
lies within the rim center plane. The transition from the point
takes place according to Eq. (3.8) by the terms
24
a ex
and
b ey
P0
normal
en .
P0 .
With an uneven
contact point.
Tire models which can be simulated within acceptable time assume that the contact patch
is suciently at. At an ordinary passenger car tire, the contact area has at normal load
approximately the size of
to fractions of millimeters, when later on the real track will be approximated by a plane
in the range of centimeters. If the track in the contact area is replaced by a local plane,
no further iterative improvements will be necessary for the contact point calculation.
ex =
eyr
eyR ezB
.
| eyR ezB |
(3.14)
xi
= yi ,
zi
i = 1(1)4 .
(3.15)
In order to sample the contact patch as good as possible the tire width
r0Qi ,0 = r0M,0 + rM Qi ,0
tire radius
r0
x r0 ex
r0 ezB ,
rM Q2 = x r0 ex
r0 ezB ,
rM Q1 =
y b eyR r0 ezB ,
rM Q3 =
(3.16)
y b eyR r0 ezB
rM Q4 =
in the front, in the rear, in the left, and in the right of the contact patch.
According to Eq. (3.5) the corresponding points on the track are given by
r0Qi ,0
xi
=
yi
,
z (xi , yi )
i = 1(1)4 .
(3.17)
The calculation of the track normal is straight forward now, Figure 3.7. The vectors
25
Vehicle Dynamics
eyR
Q1*
rMP*
en
Q1
Q3*
Q3
P*
rQ2Q1
P
Q4*
Q4
Q2*
rQ3Q4
Q2
and
en =
The unit vectors
ex , ey
rQ2 Q1 rQ4 Q3
.
| rQ2 Q1 rQ4 Q3 |
(3.18)
r0P0 ,0 =
serves as rst improvement of
point
1
(r0Q1 ,0 + r0Q2 ,0 + r0Q3 ,0 + r0Q4 ,0 )
4
(3.19)
the corresponding
in the rim center plane is obtained by Eqs. (3.12) and (3.13). On rough roads
the point
not always is located on the track. But, together with the local track normal
it represents the local road plane very well. As in reality, sharp bends and discontinuities
which will occur at step- or ramp-sized obstacles are smoothed by this approach.
=0
A0 = 4z b ,
where
(3.20)
4z = r0 rS .
(3.21)
Here, the width of the tire simply equals the width of the contact zone,
wC = b.
On a cambered tire the deected zone of the tire cross section depends on the contact
situation. The magnitude of the tire ank radii
rSL = rs +
26
b
tan
2
and
rSR = rs
b
tan
2
(3.22)
=/ 0
=0
eyR
eyR
rS
rSL
r0
en
eyR
rSR
rS
en
rS
r0
en
r0
z
b
wC = b
rSR
b*
wC
wC
full contact
partial contact
rSL r0
and
rSR r0
A =
1
(r0 rSR + r0 rSL ) b = (r0 rS ) b .
2
A0 = A
(3.23)
results in
4z = r0 rS .
Hence, at full contact the tire camber angle
(3.24)
But, due to
wC =
b
cos
(3.25)
the width of the contact area increases with the tire camber angle.
The deected zone will change to a triangular shape if one of the ank radii exceeds the
undeected tire radius. Assuming
rSL > r0
obtained by
A =
and
rSR < r0
1
(r0 rSR ) b ,
2
(3.26)
b =
r0 rSR
.
tan
(3.27)
1 (r0 rSR )2
A =
.
2 tan
(3.28)
27
Vehicle Dynamics
A0 = A
results in
1 r0 rS + 2b tan
4z =
2
b tan
where Eq. (3.22) was used to express the ank radius
tire width
wC =
2
rSR
(3.29)
| tan |
r0 rS + 2b tan
r0 rSR
b
=
=
,
cos
tan cos
sin
and
| sin |
the
(3.30)
where the Eqs. (3.27) and (3.22) where used to simplify the expression. If
are replaced by
rS ,
tan
and
sin
then, the Eqs. (3.29) and (3.30) will hold for positive
4zF
and
4zB
4z = 4zF + 4zB = r0 rS
(3.31)
will hold.
undeformed
belt
Belt
Fz
Rim
zF
rS r0
r0
L/2
zB
zB
4zF 4zB
1
4z .
2
(3.32)
Approximating the belt deection by truncating a circle with the radius of the undeformed
tire results in
28
2
L
+ (r0 4zB )2 = r02 .
2
(3.33)
4zB r0 .
L2
= 2 r0 4zB
4
L =
or
p
8 r0 4zB .
(3.34)
Inspecting the passenger car tire footprint in Figure 3.1 leads to a contact patch length of
L 140mm. For this tire the radial stiness and the inated radius are specied with cR =
265 000 N/m and r0 = 316.9 mm. The overall tire deection can be estimated by 4z =
Fz /cR . At the load of Fz = 4700N the deection amounts to 4z = 4700N / 265 000N/m =
0.0177 m. Then, by approximating the belt deformation
by the half of the tire deection,
p
the length of the contact patch will become L =
8 0.3169 m 0.0177/2 m = 0.1498 m =
150 mm which corresponds quite well with the length of the tire footprint.
Q,
FZ
to the
Figure 3.10.
rS
en
P
ey
r0-rSL
Fz
A
r0-rSR
y
wC
yQ .
By splitting the
yQ =
y A + y 4 A4
.
A
(3.35)
The minus sign takes into account that for positive camber angles the acting point will
move to the right whereas the unit vector
ey
A =
1
(r0 rSL + r0 rSR ) wC ,
2
(3.36)
29
Vehicle Dynamics
wC
A = 4z wC .
(3.37)
As the center of the rectangular section is located on the center line which runs through the
geometric contact point,
y = 0
y4 =
1
1
1
wC wC = wC .
2
3
6
(3.38)
A4 =
1
1
1
(r0 rSR (r0 rSL )) wC =
(rSL rSR )) wC =
(b tan ) wC ,
2
2
2
(3.39)
where Eq. (3.22) was used to simplify the expression. Now, Eq. (3.35) can be written as
yQ =
1
6
wC
1
2
b tan wC
b tan
b2 tan
=
wC =
.
4z wC
12 4z
12 4z cos
(3.40)
If the cambered tire has only a partial contact to the road then, according to the deection
area a triangular pressure distribution will be assumed, Figure 3.11.
b/2
en
ey
Q
Fz
y
wC
yQ =
where the width of the contact area
wC
is given by
1
b
wC
3
2 cos
,
(3.41)
b/(2 cos )
contact patch. The plus sign holds for positive and the minus sign for negative camber
angles.
The static contact point
r0Q = r0P + yQ ey
represents the contact patch much better than the geometric contact point
30
(3.42)
P.
as follows
r0Q = r0M + rM Q ,
where
(3.43)
r0M,0 = v0M,0
(3.44)
rM Q
takes part on all those motions of the wheel carrier which do not contain elements of the
wheel rotation and it In addition, it contains the tire deection
4z
rM Q,0 = 0R,0
rM Q,0 + 4z en,0 ,
where
0R
(3.45)
is the angular velocity of the wheel rim without any component in the direction
4z
en
describes
v0Q,0
(3.46)
track
eTn,0 v0P,0 = 0
As
en,0
or
is a unit vector,
eTn,0 en,0 = 1
(3.47)
(3.48)
Finally, the components of the contact point velocity in longitudinal and lateral direction
are obtained from
rM Q,0
vx = eTx,0 v0Q,0 = eTx,0 v0M,0 + 0R,0
(3.49)
and
rM Q,0 ,
vy = eTy,0 v0P,0 = eTy,0 v0M,0 + 0R,0
where the relationships
(3.50)
eTx,0 en,0 = 0 and eTy,0 en,0 = 0 were used to simplify the expressions.
31
Vehicle Dynamics
deflected tire
rigid wheel
r0 r
S
rD
vt
x
Figure 3.12: Dynamic rolling radius
r0
as unloaded and
rS = r0 4r
r0 sin 4 = x
(3.51)
r0 cos 4 = rS
(3.52)
and
hold.
If the motion of a tire is compared to the rolling of a rigid wheel, then, its radius
have to be chosen so that at an angular rotation of
rD
will
r0 sin 4 = x = rD 4 .
(3.53)
rD =
For
4 0
r0 sin 4
.
4
(3.54)
rD = r0 .
At small, yet nite angular rotations the sine-function can be approximated by the rst
terms of its Taylor-Expansion. Then, Eq. (3.54) reads as
rD
4 16 43
= r0
= r0
4
1
1 42
6
.
(3.55)
rS
1
= cos 4 = 1 42
r0
2
32
or
4 = 2
rS
1
r0
(3.56)
r D = r0
Due to
rS = rS (Fz )
1
1
3
rS
1
r0
rD
=
2
1
r0 + rS .
3
3
Fz . Therefore,
velocity , then
called dynamic tire radius. If the tire rotates with the angular
(3.57)
v t = rD
it is
(3.58)
will denote the average velocity at which the tread particles are transported through the
contact patch.
[mm]
10
0
-10
r -r
D
-20
Fz [kN]
Measurements
TMeasy tire model
FzS
= r0 + (1 ) r0
c
|
{z 0 }
rS
rD
(3.59)
where the static tire radius rS = r0 4r has been approximated by using the linearized
S
tire deformation 4r = Fz /c0 . The parameter is modeled as a function of the wheel
load Fz
= N + ( 2N N )
N and 2N
Fz = 2FzN .
where
Fz
1
FzN
Fz = FzN
(3.60)
[N/m],
[N/m],
190000.
206000.
0.375
0.750
the approximation of measured tire data can be done very well, Figure 3.13.
33
Vehicle Dynamics
Fz
4z
4z
Fz = Fz (4z, 4z)
.
(3.61)
Because the tire can only apply pressure forces to the road the normal force is restricted
to
Fz 0.
In a rst approximation
Fz
Fz = FzS + FzD .
(3.62)
The static part is described as a nonlinear function of the normal tire deection
FzS = a1 4z + a2 (4z)2 .
The constants
a1
and
a2
(3.63)
payload
cN
d FzS
=
d 4z FzS =FzN
c2N
and
d FzS
=
.
d 4z FzS =2FzN
(3.64)
d FzS
= a1 + 2 a2 4z .
d 4z
(3.65)
4z =
a1
(3.66)
Because the tire deection is always positive, the minus sign in front of the square root
has no physical meaning, and can be omitted therefore. Hence, Eq. (3.65) can be written
as
d FzS
= a1 + 2 a2
d 4z
a1 +
!
=
(3.67)
cN
c2N
or
c2N
or
c22N
and
a2 =
a21
(3.68)
8a2 FzN
nally leading to
a1 =
34
2 c2N c22N
c22N c2N
.
4 FzN
(3.69)
Results for a passenger car and a truck tire are shown in Figure 3.14. The parabolic
approximation in Eq. (3.63) ts very well to the measurements. The radial tire stiness
190 000N/m.
The Payload
80
60
40
20
2
0
100
Fz [kN]
Fz [kN]
10
10
20
30
z [mm]
40
50
20
40
60
z [mm]
80
Measurements, Approximation
FzD = dR 4z ,
where
dR
(3.70)
is a constant describing the radial tire damping, and the derivative of the tire
deformation
4z
Fz
Mx = Fz y
acting around a longitudinal axis in
P,
to the
Figure 3.15.
Note: Figure 3.15 shows a negative tipping torque. Because a positive camber angle moves
the contact point into the negative
torque.
As long as the cambered tire has full contact to the road the lateral displacement
is
Mx = Fz
b2 tan
.
12 4z cos
(3.72)
35
Vehicle Dynamics
en
ey
en
ey
P Q
y
Fz
Fz
Tx
|| 1
Mx = cN 4z
where the term
1
12
c N b2
Fz cN 4z
and small
1
b2
=
c N b2 ,
12 4z
12
(3.73)
en
ey
Q
P
Fz
represents the contact area more precisely than the geometric contact point
P.
Fz
y -axis.
If the tire rotates tread particles will be stued into the front of the contact area which
causes a slight pressure increase, Figure 3.17. Now, the resulting vertical force is applied
in front of the contact point and generates the rolling resistance torque
ty = Fz xR sign() ,
36
(3.74)
rotating
en
ex non-rotating
xR
ex
en
C
Fz
Fz
where
sign() assures
xR from C to
r0
distance
radius
that
ty
fR =
Fz
The
xR
.
r0
(3.75)
The dimensionless rolling resistance coecient slightly increases with the traveling velocity
of the vehicle
fR = fR (v) .
20 km/h < v < 200 km/h, the rolling
tires is in the range of 0.01 < fR < 0.02.
(3.76)
resistance
The rolling resistance hardly inuences the handling properties of a vehicle. But it plays
a major part in fuel consumption.
vx .
and
bed is controlled to the loaded tire radius corresponding to the wheel load
Fz , Figure 3.18.
u = (rD vx ) t .
(3.77)
T =
L
,
rD ||
(3.78)
37
Vehicle Dynamics
rD
vx
rD
vx
L
u max
where
T >0
is assured by
||.
The maximum deection occurs when the tread particle leaves the contact patch at the
time
t=T
umax = (rD vx ) T = (rD vx )
L
.
rD ||
(3.79)
The deected tread particle applies a force to the tire. In a rst approximation we get
Fxt = ctx u ,
where
ctx
(3.80)
On normal wheel loads more than one tread particle is in contact with the track, Figure
3.19a. The number
p =
where
a)
Figure 3.19:
(3.81)
b)
L
,
s+a
cxt * u
cut * u max
Particles entering the contact patch are undeformed, whereas the ones leaving have the
maximum deection. According to Eq. (3.80), this results in a linear force distribution
38
versus the contact length, Figure 3.19b. The resulting force in longitudinal direction for
particles is given by
Fx =
1 t
p c umax .
2 x
(3.82)
Fx =
1 L t
L
cx (rD vx )
.
2 s+a
rD ||
(3.83)
A rst approximation of the contact length L was calculated in Eq. (3.34). Approximating
1
F /c results in
the belt deformation by 4zB
2 z R
Fz
,
cR
L2 4 r 0
where
cR
(3.84)
denotes the radial tire stiness, and nonlinearities and dynamic parts in the tire
Fx = 2
rD v x
r0 ctx
Fz
.
s + a cR
rD ||
(3.85)
The nondimensional relation between the sliding velocity of the tread particles in lonS
gitudinal direction vx = vx rD and the average transport velocity rD || form the
longitudinal slip
sx =
The longitudinal force
sx
Fx
(vx rD )
.
rD ||
(3.86)
Fz
Fx = k Fz sx ,
where the constant
(3.87)
r0 , s, a, ctx
and
cR .
Eq. (3.87) holds only as long as all particles stick to the track. At moderate slip values
the particles at the end of the contact patch start sliding, and at high slip values only the
parts at the beginning of the contact patch still stick to the road, Figure 3.20.
L
t
Fx <= FH
adhesion
L
t
Fx = FG
Fx = FH
sliding
adhesion sliding
Fx
39
Vehicle Dynamics
Fx
M
adhesion
sliding
Fx
S
Fx dFx0
sSx
sM
x
sx
sx ,
vyS
sy =
,
rD ||
(3.88)
vyS = vy
(3.89)
vy
As long as the tread particles stick to the road (small amounts of slip), an almost linear
distribution of the forces along the length
slip values the particles at the end of the contact patch start sliding, and at high slip
values only the parts at the beginning of the contact patch stick to the road, Figure 3.22.
Fy
sliding
Fy
adhesion
sliding
L
Fy n
adhesion
The nonlinear characteristics of the lateral force versus the lateral slip can be described
0
M
M
by the initial inclination (cornering stiness) dFy , the location sy and the magnitude Fy
40
sSy ,
FyS
of the sliding
force.
The distribution of the lateral forces over the contact patch length also denes the point
of application of the resulting lateral force. At small slip values this point lies behind
the center of the contact patch (contact point P). With increasing slip values it moves
forward, sometimes even before the center of the contact patch. At extreme slip values,
when practically all particles are sliding, the resulting force is applied at the center of the
contact patch.
The resulting lateral force
Fy
as a lever
TS = n Fy .
The lateral force
sy .
Fy
(3.90)
Typical plots of these quantities are shown in Figure 3.23. Characteristic parameters
n/L
(n/L)0
Fy
M
adhesion
Fy
adhesion
adhesion/sliding
full sliding
adhesion/
sliding
s0y
full sliding
Fy dF0y
MS
sSy
sy
adhesion
adhesion/sliding
full sliding
sM
y
sSy
sy
s0y
sSy
sy
Figure 3.23: Typical plot of lateral force, tire oset and self aligning torque
M
0
of the lateral force graph are initial inclination (cornering stiness) dFy , location sy and
S
S
M
magnitude of the maximum Fy , begin of full sliding sy , and the sliding force Fy .
The dynamic tire oset has been normalized by the length of the contact patch L. The
0
S
initial value (n/L)0 as well as the slip values sy and sy suciently characterize the graph.
41
Vehicle Dynamics
Longitudinal force
Fz = 3.0 kN
Fx
Lateral force
Fz = 6.0 kN
Fz = 3.0 kN
Fy
Fz = 6.0 kN
dFx0 = 70 kN
dFx0 = 220 kN
dFy0 = 72 kN
dFy0 = 130 kN
sM
x = 0.160
sM
x = 0.120
sM
y = 0.180
sM
y = 0.200
sSx = 0.500
sSy = 0.500
sSy = 0.700
Table 3.3: Characteristic tire data with degressive wheel load inuence
friction force. With rising imperfection of the pressure distribution over the contact area,
the ability to transmit forces of friction between tire and road lessens.
In practice, this leads to a degressive inuence of the wheel load on the characteristic
curves of longitudinal and lateral forces. In order to respect this fact in a tire model, the
N
N
characteristic data for two nominal wheel loads Fz and 2 Fz are given in Table 3.3.
M
M
0
0
From this data the initial inclinations dFx , dFy , the maximal forces Fx , Fx and the
M
S
sliding forces Fx , Fy for arbitrary wheel loads Fz are calculated by quadratic functions.
For the maximum longitudinal force it reads as
Fz
1 M
1 M
N
M
N Fz
N
N
M
= N 2 Fx (Fz ) 2 Fx (2Fz ) Fx (Fz ) 2 Fx (2Fz ) N .
Fz
Fz
Fy [kN]
Fx [kN]
FxM (Fz )
-2
-2
-4
-4
-6
-0.4
-0.2
0.2
0.4
-6
-20
-10
M
sM
x , sy ,
10
20
42
[deg]
sx [-]
(3.91)
sM
x (Fz )
N
sM
x (Fz )
N
sM
x (2Fz )
F
z
1 .
FzN
N
sM
x (Fz )
(3.92)
With the numeric values from Tab. 3.3 a slight shift of the maxima with an increasing
wheel load is also modeled, Figure 3.24.
Fx = Fx (sx )
force depending on the lateral slip Fy = Fy (sy ) can be dened by their characteristic
0
0
M
M
parameters initial inclination dFx , dFy , location sx , sy and magnitude of the maximum
M
M
S
S
Fx , Fy as well as sliding limit sx , sy and sliding force FxS , FyS , Figure 3.25. During
general driving situations, e.g. acceleration or deceleration in curves, longitudinal sx and
sy
lateral slip
Fx
appear simultaneously.
Fx
Fx
Fy
0
dF x
sx
sSx
sM
x
Fy Fy
dF 0
FS
FM
Fy
F(s)
Fx
sy
dF y
sSy
sS
sy
sM
sM
y
sx
sx
sy
slip
s
s =
where the slips
sx
and
sy
sx
sx
2
+
sy
sy
2
,
(3.93)
sx
and
sy .
43
Vehicle Dynamics
Similar to the graphs of the longitudinal and lateral forces the graph of the generalized tire
0
M
M
S
S
force is dened by the characteristic parameters dF , s , F , s and F . The parameters
are calculated from the corresponding values of the longitudinal and lateral force
dF
sM =
FM =
sS =
FS =
q
2
(dFx0 sx cos )2 + dFy0 sy sin ,
s
2 M
2
sy
sM
x
cos +
sin ,
sx
sy
q
2
(FxM cos )2 + FyM sin ,
s
2 S
2
sy
sSx
cos +
sin ,
sx
sy
q
2
(FxS cos )2 + FyS sin ,
(3.94)
where the slip normalization have also to be considered at the initial inclination. The
angular functions
cos =
sx /
sx
s
and
sin =
sy /
sy
s
(3.95)
grant a smooth transition from the characteristic curve of longitudinal to the curve of
,
sM dF 0
= M ,
0 s sM ;
s
s
1 + + dF 0 M 2
F
F (s) =
(3.96)
s sM
M
M
S
2
M
S
(F
F
)
(3
2
)
,
=
,
s
<
s
s
;
sS sM
FS ,
s > sS .
The function
0
When dening the curve parameters, one just has to make sure that the condition dF
M
2 FsM is fullled, because otherwise the function has a turning point in the interval 0
s sM .
<
Now, the longitudinal and the lateral force follow from the according projections in longitudinal and lateral direction
Fx = F cos
and
Fy = F sin .
(3.97)
44
30
3
20
10
F [kN]
1
0
F [kN]
-1
-10
-2
-20
-3
-4
-2
0
F [kN]
-30
-20
0
F [kN]
20
|| = 1, 2, 4, 6, 10, 14
|sx | = 1, 2, 4, 6, 10, 15 %;
Fz = 3.2 kN
Fz = 35 kN
Fy [N]
4000
L /0
3000
0.2
2000
1000
0.4
0.6
-1000
0.8
-2000
1.0
-3000
-4000
-0.5
sy [-]
0.5
Fz = 3.2 kN
sM
L M
L M
s , FM
F ,
0
0
sS
L S
L S
s , FS
F ,
0
0
(3.98)
45
Vehicle Dynamics
the essential tire model parameter which are valid for the friction coecient
justed to the new friction coecient
L .
are ad-
Figure 3.27.
If the road model provides not only the roughness information
local friction coecient
[z, L ] = fR (x, y)
then, braking on
z = fR (x, y)
-split
be simulated.
nN = n/L. It mainly depends on the lateral slip sy . The normalized tire oset starts
at sy = 0 with an initial value (n/L)0 . It tends to zero, n/L 0 at large slip values,
sy sSy . Sometimes the dynamic tire oset overshoots to negative values before it reaches
S
0
zero again. This behavior can be modeled by introducing the parameter sy < sy , Figure
3.28.
n/L
n/L
(n/L)0
(n/L)0
s0y
sy
sSy
s0y
sy
In order to achieve a simple and smooth approximation of the normalized tire oset versus
0
the lateral slip, a linear and a cubic function are overlayed in the rst section sy sy
|sy |
sy
2
0 S
n n
|sy | sy sy |sy |
=
(1w)
s0y < |sy | sSy
L
L 0
0
S s0
s
s
y
y
y
0
|sy | > sSy
where the factor
s0y
w= S
sy
(3.99)
(3.100)
0
weights the linear and the cubic function according to the values of the parameter sy
S
0
S
and sy . No overshoot occurs for sy = sy . Here, w = 1 and (1 w) = 0 will produce a
cubic transition from
sy = s0y .
46
n/L = (n/L)0
to
The characteristic curve parameters, which are used for the description of the dynamic
tire oset, are at rst approximation not wheel load dependent. Similar to the description
of the characteristic curves of longitudinal and lateral force, here also the parameters for
single and double pay load are given. The calculation of the parameters of arbitrary wheel
loads is done similar to Eq. (3.92) by linear inter- or extrapolation.
Tz [Nm]
200
150
100
Fz
50
Fz = 3.0 kN
Fz = 6.0 kN
0
-50
s0y = 0.200
s0y = 0.220
-100
sE
y = 0.420
sE
y = 0.400
-150
-200
-30
(n/L)0
-20
-10
10
20
[deg]
30
sy 0
one
gets at rst approximation a triangular distribution of lateral forces over the contact area
length cf. Figure 3.22. The working point of the resulting force (dynamic tire oset) is
then given by
n(Fz 0, sy = 0) =
The value
1
L.
6
(3.101)
n = 16 L can only serve as reference point, for the uneven distribution of pressure
in longitudinal direction of the contact area results in a change of the deexion prole
and the dynamic tire oset.
The self aligning torque in Figure 3.29 has been calculated with the tire parameter from
Table 3.3. The degressive inuence of the wheel load on the self aligning torque can be
seen here as well.
With the parameters for the description of the tire oset it has been assumed that at
N
the payload Fz = Fz the related tire oset reaches the value of (n/L)0 = 0.17 1/6 at
sy = 0. The slip value s0y , at which the tire oset passes the x-axis, has been estimated.
Usually the value is somewhat higher than the position of the lateral force maximum.
S
With rising wheel load it moves to higher values. The values for sy are estimated too.
47
Vehicle Dynamics
has a component
n = sin .
en
rim
centre
plane
eyR
(3.102)
Fy = Fy (s y ): Parameter
4000
3000
2000
1000
0
-1000
ex
ey
rD ||
y()
v()
-2000
-3000
-4000
-0.5
Fy ()
at
0.5
Fz = 3.2 kN
and
= 0 , 2 , 4 , 6 , 8
Now, the tread particles in the contact patch possess a lateral velocity which depends on
their position
and is provided by
v () = n
L
, = sin ,
2 L/2
L/2 L/2 .
(3.103)
At the contact point it vanishes whereas at the end of the contact patch it takes on the
same value as at the beginning, however, pointing into the opposite direction. Assuming
that the tread particles stick to the track, the deection prole is dened by
y () = v () .
(3.104)
y () =
where
rD ||
(3.105)
denotes the average transport velocity. Now, Eq. (3.104) can be written as
d y ()
rD || = sin
d
48
d y () d
d y ()
=
rD ||
d dt
d
or
d y ()
sin L
=
,
d
rD || 2 L/2
(3.106)
where
L/2
sy
which
sin L
.
rD || 2
(3.107)
d y ()
= s
.
d
L/2
(3.108)
s =
Then, Eq. (3.106) simplies to
2
1 L
+ C.
(3.109)
y = s
2 2 L/2
1
The boundary condition y = L = 0 can be used to determine the integration constant
2
C . One gets
1 L
C = s
.
(3.110)
2 2
Then, Eq. (3.109) reads as
1 L
y () = s
2 2
"
1
L/2
2 #
.
(3.111)
The lateral displacements of the tread particles caused by a camber slip are compared
now with the ones caused by pure lateral slip, Figure 3.31. At a tire with pure lateral
a) camber slip
y
b) lateral slip
yy()
y()
_
yy
_
y
-L/2
L/2
-L/2
L/2
dyy /d rD || = vy ,
y y was
dyy /d . Hence, the deection prole is linear, and
reads as yy = vy /(rD ||) = sy , where the denition in Eq. (3.88) was used to
introduce the lateral slip sy . Then, the average deection of the tread particles under
results in
yy = sy
L
.
2
(3.112)
49
Vehicle Dynamics
The average deection of the tread particles under pure camber slip is obtained from
ZL/2 "
1 L 1
y = s
2 2 L
1
x
L/2
2 #
1
L
d = s .
3
2
(3.113)
L/2
A comparison of Eq. (3.112) with Eq. (3.113) shows, that by using
1
s
3
sy =
the lateral camber slip
(3.114)
sy .
In normal driving conditions, the camber angle and thus, the lateral camber slip are
limited to small values. So, the lateral camber force can be approximated by
Fy dFy0 sy .
If the global inclination
dFy = Fy /sy
(3.115)
dFy0 ,
one
gets the camber inuence on the lateral force as shown in Figure 3.30.
The camber angle inuences the distribution of pressure in the lateral direction of the
contact patch, and changes the shape of the contact patch from rectangular to trapezoidal. Thus, it is extremely dicult, if not impossible, to quantify the camber inuence
with the aid of such simple models. But, it turns out that this approach is quit a good
approximation.
0W = 0R
+ eyR
has a component in direction of the track normal
(3.116)
en
n = eTn 0W 6= 0 .
(3.117)
Then, a very complicated deection prole of the tread particles in the contact patch
occurs. However, by a simple approach the resulting bore torque can be approximated
quite good by the parameter of the generalized tire force characteristics.
At rst, the complex shape of a tire's contact patch is approximated by a circle, Figure
3.32. By setting
1
R =
2
L B
+
2
2
1
(L + B)
4
length L and
(3.118)
the width
of the actual
contact patch. The integration over the whole circle area results in the bore torque
TB
50
1
=
A
Z
F r dA ,
A
(3.119)
ex
B
dr
F
r
L
ey
circular
approximation
TB
1
= 2
R
ZR Z2
F r rd dr
0
(3.120)
TB
R
2
Z
ZR
1
2
F r r dr = 2
= 2
F r2 dr .
R
R
0
(3.121)
For small slip values the force transmitted in the patch element can be approximated by
F = F (s) dF 0 s
where
dF 0
(3.122)
is the initial inclination of the
generalized tire force characteristics. Similar to Eqs. (3.86 and (3.88) we dene
s =
where
r n
r n
rD ||
(3.123)
rD
and
denote the
TB
2
= 2
R
ZR
dF 0
r n 2
r dr
rD ||
(3.124)
51
Vehicle Dynamics
TB
2
n
= 2 dF 0
R
rD ||
ZR
2
1
R n
n R4
0
r dr = 2 dF
=
R dF 0
R
rD || 4
2
rD ||
3
(3.125)
0
where
sB =
R n
rD ||
(3.126)
dF 0
TB
takes
TBmax
2
= 2
R
ZR
F S r2 dr =
2 S R3
2 S
=
F R.
F
R2
3
3
(3.127)
0
Due to the generalized sliding force
FS
TBmax
depends on the
tire properties and the actual friction value. Now, the bore torque is given by
TB =
1
R n
R dF 0
2
rD ||
|TB |
with
2 S
F R
3
(3.128)
and
truck tire
40
20
F [kN]
2
0
F [kN]
1.8 kN
3.2 kN
4.6 kN
5.4 kN
-2
-4
-6
-40
10 kN
20 kN
30 kN
40 kN
50 kN
-20
-40
-20
20
40
-40
-20
sx [%]
Figure 3.33: Longitudinal force:
Meas.,
sx [%]
20
40
TMeasy
The tire model TMeasy which is based on this approach, can be used for passenger car
tires as well as for truck tires. It approximates the characteristic curves
52
Fx = Fx (sx ),
passenger car
truck
40
Fy [kN]
F [kN]
20
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-2
-4
-40
100
1000
50
500
[Nm]
1500
[Nm]
150
-50
1.8 kN
3.2 kN
4.6 kN
6.0 kN
-100
-150
-20
-10
[o]
10
20
10 kN
20 kN
30 kN
40 kN
-20
-6
Mz
-500
18.4 kN
36.8 kN
55.2 kN
-1000
-1500
-20
Fy = Fy ()
and
Mz = Mz ()
-10
[o]
Meas.,
10
20
TMeasy
Fz ,
Figures 3.33
and 3.34.
When experimental tire values are missing, the model parameters can be pragmatically
estimated by adjustment of the data of similar tire types. Furthermore, due to their physical signicance, the parameters can subsequently be improved by means of comparisons
between the simulation and vehicle testing results as far as they are available.
53
4 Suspension System
4.1 Purpose and Components
The automotive industry uses dierent kinds of wheel/axle suspension systems. Important
criteria are costs, space requirements, kinematic properties, and compliance attributes.
The main purposes of a vehicle suspension system are
guiding elements:
control arms, links,
struts,
leaf springs,
force elements:
coil spring, torsion bar, air spring, leaf spring,
anti-roll bar,
damper,
bushings, hydro-mounts,
tires.
Tires are air springs that support the total weight of the vehicle. The air spring action of
the tire is very important to the ride quality and safe handling of the vehicle.
54
leaf springs
links
55
Vehicle Dynamics
The semi-trailing arm is a simple and cheap design which requires only few space. It is
mostly used for driven rear axles.
The short-long-arm axle design allows a nearly independent layout of longitudinal and
lateral axle motions. It is similar to the central control arm axle suspension, where the
trailing arm is completely rigid and hence, only two lateral links are needed.
The twist beam axle suspension exhibits either a trailing arm or a semi-trailing arm
characteristic. It is used for non driven rear axles only. The twist beam axle provides
enough space for spare tire and fuel tank.
56
30
must
be provided at the front wheels of passenger cars. Depending on the wheel base, busses
uR S
nk
drag li
wheel
and
wheel
body
rack
pinion
steering
box
overall steering ratio depends on the ratio of the steering box and on the kinematics of
the steering linkage.
ring
ever
l
ring
stee
lev
er 1
drag link 1
drag link 2
2
wheel and
wheel body
Figure 4.5: Lever arm steering system
Using a lever arm steering system Fig. 4.5, large steering angles at the wheels are possible.
This steering system is used on trucks with large wheel bases and independent wheel
57
Vehicle Dynamics
suspension at the front axle. Here, the steering box can be placed outside of the axle
center.
Firstly, the rotations of the steering wheel
rotation of the steer levers
1 = 1 (L ), 2 = 2 (L ).
L
wheel
and
wheel
body
g
steerin
r
leve
steer box
O
(90o rotated)
steering link
2
drag link
Figure 4.6: Drag link steering system
S are transformed by the steering box to the rotation of the steering lever arm
L = L (S ) and further on to the rotation of the left wheel, 1 = 1 (L ). The drag link
transmits the rotation of the left wheel to the right wheel, 2 = 2 (1 ). The steering ratio
wheel
is dened by the ratio of the steering box and the kinematics of the steering link. Here,
the ratio
2 = 2 (1 )
angles at the front wheels are needed to achieve a good manoeuvrability. That is why,
more sophisticated steering systems are needed, Fig. 4.7. The rotations of the steering
S are transformed by the steering box to the rotation of the steering lever arm
L = L (S ). The left lever arm is moved via the steering link A = A (L ). This motion
wheel
is transferred by a coupling link to the right lever arm. Finally, the left and right wheels
are rotated via the drag links,
58
1 = 1 (A )
and
2 = 2 (A ).
steer
in
g leve
r
L
steering box
steering link
left
lever
arm
drag link
coupl.
link
2
wheel and
wheel body
Coil spring
FS
u
u
FS
Air spring
Torsion bar
Leaf spring
FS
FS
Figure 4.8: Vehicle suspension springs
Coil springs, torsion bars, and leaf springs absorb additional load by compressing. Thus,
the ride height depends on the loading condition. Air springs are rubber cylinders lled
with compressed air. They are becoming more popular on passenger cars, light trucks, and
heavy trucks because here the correct vehicle ride height can be maintained regardless of
the loading condition by adjusting the air pressure.
59
Vehicle Dynamics
FS
FS
u
LF
FS0
L
c
L0
FS
Figure 4.9: Linear coil spring and general spring characteristics
LF
c,
FS = c LF L ,
where
(4.1)
L denotes the actual length of the spring. Mounted in a vehicle suspension the spring
has to support the corresponding chassis weight. Hence, the spring will be compressed to
the conguration length
L0 < L F .
FS = c LF (L0 u)
where
FS0
= c LF L0 + c u = FS0 + c u ,
(4.2)
FS
u
FS = FS (u) .
(4.3)
FS = FS (u).
4.4.2 Damper
Dampers are basically oil pumps, Fig. 4.10. As the suspension travels up and down, the
hydraulic uid is forced by a piston through tiny holes, called orices. This slows down
the suspension movement.
60
FD
Rebound Ch.
Piston
FD
Compression
Chamber
Piston orifice
v
Figure 4.10: Principle of a mono-tube damper
Today twin-tube and mono-tube dampers are used in vehicle suspension systems. Dynamic
damper models compute the damper force via the uid pressure applied to each side of
the piston. The change in uid pressures in the compression and rebound chambers are
calculated by applying the conservation of mass.
In standard vehicle dynamics applications simple characteristics
FD = FD (v)
are used to describe the damper force
FD
(4.4)
v.
To
obtain this characteristics the damper is excited with a sinusoidal displacement signal
FD = FD(u)
FD = FD(v)
FD [N]
1000
0
Compression
f0
-1000
-2000
Rebound
-3000
fE
-4000
-0.06 -0.04 -0.02
u [m]
-1.2
-0.8
-0.4
0.4
0.8
1.2
1.6
v [m/s]
61
Vehicle Dynamics
Topmount
Stop
Control arm
bushings
Subframe mounts
properties change with the frequency of the motion. Here, more sophisticated dynamic
models are needed.
62
xe (t) = A sin(2 f t) ,
with dierent frequencies
f0 f fE
(4.5)
and amplitudes
Amin A Amax .
Starting at
t = 0,
the system will usually be in a steady state condition after several periods
where
T = 1/f
and
n = 2, 3, . . .
t nT ,
n T t (n + 1)T ,
sin(2 f t) + cos(2 f t) .
F (t)
|
{z
}
|{z}
rst harmonic approximation
measured/
calculated
The coecients
1
2
(n+1)T
Z
and
(4.6)
2
sin(2 f t)+ cos(2 f t) F (t) dt
M inimum .
(4.7)
nT
The dierentiation of Eq. (4.7) with respect to
and
necessary conditions
(n+1)T
Z
nT
(n+1)T
Z
sin(2 f t)+ cos(2 f t) F (t) sin(2 f t) dt = 0
(4.8)
sin(2 f t)+ cos(2 f t) F (t) cos(2 f t) dt = 0
nT
with the solutions
R
R
R
F sin dt cos2 dt F cos dt sin cos dt
R
R
R
=
sin2 dt cos2 dt 2 sin cos dt
,
R
R
R
R
F cos dt sin2 dt F sin dt sin cos dt
R
R
R
=
sin2 dt cos2 dt 2 sin cos dt
(4.9)
where the integral limits and arguments of sine and cosine no longer have been written.
Because it is integrated exactly over one period
nT t (n + 1)T ,
Eq. (4.9)
sin cos dt = 0 ;
sin2 dt =
R
T
T
;
cos2 dt =
2
2
(4.10)
63
Vehicle Dynamics
2
=
T
2
=
T
F sin dt ,
Z
F cos dt
(4.11)
remains. However, these are exactly the rst two coecients of a Fourier-"-Approximation.
The rst order harmonic approximation in Eq. (4.6) can now be written as
F (t) = F sin (2 f t + )
where amplitude
F =
(4.12)
are given by
2 + 2
tan =
and
(4.13)
(4.14)
F = A
c2 + (2f d)2
tan =
and
c 6= 0
and
d=0
d
d 2f A
= 2f .
cA
c
(4.15)
F = A c and
d 6= 0 results in
is characterized by
is also called
the dissipation angle can be used to evaluate the damping properties of the force element.
The dynamic stiness, dened by
cdyn =
F
A
(4.16)
xe (t) = A sin(2 f t) ,
where the period
T = 1/f
(4.17)
x e (t) = A 2 f cos(2 f t) =
2
A cos(2 f t) .
T
(4.18)
64
(4.19)
x e (t) = A 2 h(t)
cos(2 h(t)) .
In the following we demand that the function
time, i.e:
h(t)
=
where
p>0
and
q>0
1
1
=
,
T (t)
pqt
(4.21)
h(t) =
The initial condition
h(t)
(4.20)
h(t = 0) = 0
1
ln(p q t) + C .
q
(4.22)
C =
1
ln p .
q
(4.23)
With Eqs. (4.23) and (4.22) Eq. (4.19) results in a sine-like function
xe (t) = A sin
2
q
ln
p
,
pqt
(4.24)
1
p
ln
= 0, 1, 2,
q
p q tn
and
tn =
or
p
= en q ,
p q tn
mit
n = 0, 1, 2,
p
(1 en q ) , n = 0, 1, 2, .
q
(4.25)
(4.26)
The time dierence between two zero values yields the period
p
Tn = tn+1 tn = (1e(n+1) q 1+en q )
q
, n = 0, 1, 2, .
p n q
q
Tn =
e
(1 e )
q
For the rst
(n = 0)
and last
(n = N )
(4.27)
p
(1 eq )
q
.
p
=
(1 eq ) eN q = T0 eN q
q
T0 =
TN
q/p
f0
(4.28)
fE ,
1
fE
q =
ln
,
N
f0
n
hf iN o
q
E
= f0 1
,
p
f0
(4.29)
65
Vehicle Dynamics
with
xing the number of frequency intervals. The passing of the whole frequency range
1 e(N +1) q
.
q/p
tN +1 =
(4.30)
N = 500
728 s
or
0.1Hz to f = 100Hz
12min.
and
s.
Then, the the forces acting in the spring and damper are given by
FS = c s
FD = FS
displacement s
and
FD = d (u s)
.
d (u s)
= cs
or
(4.31)
d
d
s = s + u ,
c
c
d
(4.32)
acts as time
T = d/c. Hence, this force element will responds dynamically to any excitation.
u(t) = u0 sin t
respectively
u = u0 cos t
(4.33)
can be calculated easily. The steady state response will be of the same type as the excitation. Inserting
(4.34)
d
d
u0 (a cos t b sin t) = u0 (a sin t + b cos t) +
u0 cos t .
|
{z
}
|
{z
}
c
c | {z }
s
u
s
66
(4.35)
d
u0 b = u0 a
c
d
d
u0 a = u0 b +
u0
c
c
and
(4.36)
2
2 + (c/d)2
a =
and
b =
.
2
d + (c/d)2
(4.37)
FS = c s
c
sin t +
= c u0 2
cos t
d
+ (c/d)2
(4.38)
FS = FS sin (t + )
where the force magnitude
FS =
FS
p
c u0
c u0
2 + (c/d)2 = p
2
2
2
+ (c/d)
+ (c/d)2
cdyn = FS /u0
(4.39)
are given by
and
= arctan
c/d
.
(4.40)
cdyn
400
4
300
[N/mm]
200
c = 400 N/mm
2
1
100
0
100
d
d1 = 1000 N/(m/s)
d2 = 2000 N/(m/s)
[o]
50
d1 = 3000 N/(m/s)
d2 = 4000 N/(m/s)
3
4
0
20
40
60
80
100
f [Hz]
Figure 4.14: Frequency response of a spring damper combination
With increasing frequency the spring damper combination changes from a pure damper
d.
67
Vehicle Dynamics
parallel force elements, Fig. 4.15. The static load is carried by a single spring with
the stiness
c0
F0 = F0 (u).
u
FM
F1
d1
d2
FM
F2
dN
FM
FN
c0
s2
s1
c1
sN
cN
c2
FSi = ci si
were
and
si
and
are given by
FDi = di (s i u)
,
(4.41)
As long as the absolute value of the spring force FSi is lower than the maximum friction
M
force FF the damper friction combination will not move at all
u s i = 0
for
|FSi | FFM .
(4.42)
(4.43)
di (s i u)
= FSi FFM
(4.44)
F + FFM
Si
di s i =
di u
F F M
Si
68
(4.45)
FSi = ci si .
In extension to this linear approach nonlinear springs and dampers may be used. To derive
all the parameters an extensive set of static and dynamic measurements is needed.
4.5.3.1 Hydro-Mount
For the elastic suspension of engines in vehicles very often specially developed hydromounts are used. The dynamic nonlinear behavior of these components guarantees a
good acoustic decoupling but simultaneously provides sucient damping.
xe
main spring
chamber 1
membrane
cF
cT
__
2
ring channel
chamber 2
uF
c__
T
2
MF
dF
__
2
dF
__
2
uF
FH = cT xe + FF (xe uF )
MF ,
(4.46)
holds, where the force eect of the main spring has been approximated by a linear spring
with the constant
cT .
69
Vehicle Dynamics
With
MF R
as the actual mass in the ring channel and the cross sections
A K , AR
of
MF =
A 2
K
AR
MF R .
(4.47)
The uid in chamber 1 is not being compressed, unless the membrane can evade no longer.
With the uid stiness
FF (xe uF ) =
c
(x
u
)
+
s
(xe uF ) < sF
F
e
F
F
0
c (x u ) s
F
e
F
F
FF 6= 0
for
|xe uf |
sF
(4.48)
is not realistic and leads to problems, even with the numeric solution.
|xe uf | 2 sF .
The motions of the uid mass cause friction losses in the ring channel, which are as a rst
approximation proportional to the speed,
FD = dF u F .
(4.49)
MF uF = FF FD .
(4.50)
The membrane clearance makes Eq. (4.50) nonlinear and only solvable by numerical integration. The nonlinearity also aects the overall force in the hydro-mount, Eq. (4.46).
The dynamic stiness and the dissipation angle of a hydro-mount are displayed in Fig. 4.17
versus the frequency.
The simulation is based on the following system parameters
mF
25 kg
cT
dF
750 N/(m/s)
damping constant
cF
uid stiness
sF
0.0002 mm
By the nonlinear and dynamic behavior a very good compromise can be achieved between
noise isolation and vibration damping.
70
400
300
200
100
0
60
50
40
30
20
10
0
10
10
Figure 4.17: Dynamic stiness [N/mm] and dissipation angle [deg] for a hydro-mount
71
5 Vertical Dynamics
5.1 Goals
The aim of vertical dynamics is the tuning of body suspension and damping to guarantee
good ride comfort, resp. a minimal stress of the load at sucient safety.
The stress of the load can be judged fairly well by maximal or integral values of the body
accelerations.
The wheel load
Fz
Fx
and
Fy
Fz
on
Fx
Fx
and
Fy
Fy
by the coecient
72
Ford
0.000000
ZZ
Y
Y
X
X
Thilo Seibert
Ext. 37598
Vehicle Dynamics, Ford Research Center Aachen
/export/ford/dffa089/u/tseiber1/vedyna/work/results/mview.mvw
07/02/98
AA/FFA
Much simpler models can be used, however, for basic studies on ride comfort and ride
safety. A two-dimensional vehicle model, for instance, suits perfectly with a single track
road model, Fig. 5.2. Neglecting longitudinal accelerations, the vehicle chassis only per-
a2
a1
M2
zC1
M1
M*
zB
zA2
m2
M,
zR(s-a2)
zA1
zR(s+a1)
zC2
pitch
m1
hub
C
yB
xB
zR(s)
Figure 5.2: Vehicle model for basic comfort and safety analysis
forms hub and pitch motions. Here, the chassis is considered as one rigid body. Then,
mass and inertia properties can be represented by three point masses which are located in
the chassis center of gravity and on top of the front and the rear axle. The lumped mass
model has 4 degrees of freedom. The hub and pitch motion of the chassis are represented
by the vertical motions of the chassis in the front
zA1
and
zA2
zC1
the vehicle and measured at the chassis center of gravity. Then, the irregularities at the
front and rear axle are given by
zR (s a2 )
gravity C .
zR (s + a1 )
and
respectively, where
a1
and
a2
73
Vehicle Dynamics
M1 + M + M2 = M
(5.1)
and they have to provide the same inertia around an axis located in the chassis center
a21 M1 + a22 M2 = .
(5.2)
a1 M 1 = a2 M 2 .
(5.3)
M1 =
a1 (a1 +a2 )
and
M2 =
,
a2 (a1 +a2 )
(5.4)
M = M
1
M a1 a2
(5.5)
= M a1 a2 then,
the coupling mass vanishes M = 0, and the vehicle can be represented by two uncoupled
If the mass and the inertia properties of a real vehicle happen to result in
two mass systems describing the vertical motion of the axle and the hub motion of the
chassis mass on top of each axle.
mid
full
sports
size
size
utility
car
car
vehicle
m1 [kg]
80
100
125
120
600
m2 [kg]
80
100
125
180
1100
a1 [m]
a2 [m]
1.10
1.40
1.40
1.40
1.45
1.38
1.90
1.40
2.90
1.90
M [kg]
1100
1400
1950
3200
14300
[kg m2 ]
1500
2350
3750
5800
50000
M1
M [kg]
M2
545
126
429
600
200
600
914
76
960
925
1020
1255
3592
5225
5483
vehicles
properties
front axle
mass
rear axle
mass
center
of
gravity
chassis
mass
chassis
inertia
lumped
mass
model
commercial
heavy
vehicle
truck
74
Depending on the actual mass and inertia properties the vertical dynamics of a vehicle
can be investigated by two simple decoupled mass models describing the vibrations of the
front and rear axle and the corresponding chassis masses. By using half of the chassis and
half of the axle mass we nally end up in quarter car models.
The data in Table 5.1 show that for a wide range of passenger cars the coupling mass is
smaller than the corresponding chassis masses,
M < M1
and
M < M2 .
mass model or the quarter car model represent a quite good approximation to the lumped
mass model. For commercial vehicles and trucks, where the coupling mass has the same
magnitude as the corresponding chassis masses, the quarter car model serves for basic
studies only.
mi Mi , i = 1, 2.
can be investigated independently. The quarter car model is now further simplied to two
single mass models, Fig. 5.3.
``
6
zC
M
`
cS ``
`
`
`
cS `
`
dS
dS
6
zW
`
c
6
zR
``
`` c
`` T
c
6
zR
M zC + dS zC + cS zC = dS zR + cS zR
(5.6)
m zW + dS zW + (cS + cT ) zW = cT zR ,
(5.7)
and
where
zC
and
zW
label the vertical motions of the corresponding chassis mass and the
wheel mass with respect to the steady state position. The constants
cS , dS
describe the
FTD = cT (zR zW )
(5.8)
75
Vehicle Dynamics
where
cT
zR
In this simple approach the damping eects in the tire are not taken into account.
zR = 0
zR = 0.
and
z + 2 0 z + 02 z = 0 ,
(5.9)
Chassis:
Wheel:
z zC ,
C =
z zW ,
W =
d
S
,
2 cS M
dS
2
(cS +cT )m
2
02 0C
=
2
02 0W
=
cS
;
M
cS +cT
.
m
(5.10)
z(t) = z0 et ,
where
z0
and
(5.11)
(2 + 2 0 + 02 ) z0 et = 0 .
Non-trivial solutions
z0 6= 0
(5.12)
are possible, if
2 + 2 0 + 02 = 0
will hold. The roots of the characteristic equation (5.13) depend on the value of
<1 :
1 :
p
1,2 = 0 pm i 0 1 2 ,
p
2
1,2 = 0 1 .
(5.13)
(5.14)
Figure 5.4 shows the root locus of the eigenvalues for dierent values of the viscous
damping rate
76
=0
=0.5
=0.2
=0.7
-3
=1.25
-2.5
-2
-1
-1.5
-0.5
=0.9
=0.2
=0.5
Re()/0
-0.5
=0.7
For
1.0
=1.25 =1.5
=1
Im()/0
0.5
=0.9
=1.5
and
=0
-1.0
are both real and negative. Hence, Eq. (5.11) will produce a
is the
complex,
(5.15)
A and are constants which have to be adjusted to given initial conditions z(0) = z0
and z(0)
where
= 0
tends to zero,
0,
1 2
(5.16)
1.
In a more general way the relative damping may be judged by the ratio
Re(1,2 )
.
| 1,2 |
D =
(5.17)
D =
(5.18)
| 1,2 | =
r
Re(1,2 )2 + Im(1,2 )2 =
p
2
( 0 )2 + 0 1 2
= 0 ,
(5.19)
D =
For
+ 0
= .
0
(5.20)
1 the eigenvalues become real and negative. Then, Eq. (5.17) will always produce
D = 1. In this case the viscous damping rate is more
77
Vehicle Dynamics
u0 =
Mg
.
cS
(5.21)
4M g
.
cS
(5.22)
4u =
If for the maximum load variation
M M + 4M
u0 u0 + 4u.
4M
4u
cS
4M g
.
4u
(5.23)
In the standard design of a passenger car the engine is located in the front and the trunk
in the rear part of the vehicle. Hence, most of the load is supported by the rear axle
suspension.
For an example we assume that
150 kg
wheel by
umax 0.8 m
to
is in the
that the
spring deection caused by the load should not exceed half of the maximum value. Then,
according to Eq. (5.23) a lower bound of the spring rate at the front axle can be estimated
by
cmin
= ( 75 kg 9.81 m/s2 )/(0.08/2) m = 18400 N/m .
S
(5.24)
The maximum load over one rear wheel was estimated here by
that the suspension travel at the rear axle is slightly larger,
spring rate at the rear axle can be estimated by
cmin
= ( 175 kg 9.81 m/s2 )/(0.10/2) m = 34300 N/m ,
S
(5.25)
which is nearly two times the minimum value of the spring rate at the front axle. In order
to reduce this dierence we will choose a spring rate of
In Tab. 5.1 the lumped mass chassis model of a full size passenger car is described by
M1 = M2 = 600 kg
and
M = 200.
decoupled two mass models we have to neglect the coupling mass or, in order to achieve
78
M = M1 + M /2 /2 = (600 kg + 200/2 kg)/2 = 350 kg .
(5.26)
According to Eq. 5.10 the undamped natural eigen frequency of the simple chassis model
2
is then given by 0C = cS /M . Hence, for a spring rate of cS = 20000 N/m the undamped
natural frequency of the unloaded car amounts to
f0C =
(5.27)
which is a typical value for most of all passenger cars. Due to the small amount of load
the undamped natural frequency for the loaded car does not change very much,
f0C =
p
20000 N/m (350 + 75) kg/(2 ) = 1.1 Hz .
(5.28)
The corresponding cassis mass over the rear axle is given here by
M = M2 + M /2 /2 = (600 kg + 200/2 kg)/2 = 350 kg .
(5.29)
0
f0C
=
p
34300 N/m/350 kg/(2 ) = 1.6 Hz
(5.30)
L
f0C
=
(5.31)
u0
function
FS = FS0 + c0 u
where
FS0
cS
1+k
u
4u
2 !
,
(5.32)
u = 0,
and
value
FS0 + c0 4u (1 + k) = FS0 + cS 4u
must hold, where
cS
or
c0 (1 + k) = cS
(5.33)
dFS
= c0
du
1 + 3k
u
4u
2 !
.
(5.34)
79
Vehicle Dynamics
FS
dFS
du
8000
u=u
FS [N]
6000
M g
63 kN/m
44 kN/m
cS
FS0
dFS
du
4000
29 kN/m
u=0
20 kN/m
2000
0.05
u [m]
0.1
u = 4u
is given by
cL = c0 (1 + 3 k) .
The intensity of the nonlinearity
(5.35)
c0 = 20000 N/m
and rear axle will be the same for the unloaded vehicle. With
yields
k =
cS = 34300 N/m
Eq. (5.33)
34300
cS
1 = 0.715 .
1 =
c0
20000
(5.36)
The solid line in Fig. 5.5 shows the resulting nonlinear spring characteristics which is
characterized by the spring rates
c0 = 20 000 N/m
and
cL = c0 (1 + 3k) = 20 000 (1 +
for the unloaded and the loaded vehicle. Again, the undamped
natural frequencies
s
0
f0C
=
20000 N/m 1
= 1.20 Hz
350 kg
2
s
or
L
f0C
=
92000 N/m 1
= 1.74 Hz
(350+175) kg 2
(5.37)
for the unloaded and the loaded vehicle dier quite a lot.
The unloaded and the loaded vehicle have the same undamped natural frequencies if
c0
cL
=
M
M + 4M
or
cL
M + 4M
=
c0
M
(5.38)
will hold. Combing this relationship with Eq. (5.35) one obtains
M
1 + 3k =
M + 4M
80
or
1
k =
3
M + 4M
1 4M
1 =
.
M
3 M
(5.39)
cS = 34300 N/m will produce the dotted line in Fig. 5.5. The spring rates
c0 = cS /(1 + k) = 34 300 N/m / (1 + 0.1667) = 29 400 N/m for the unloaded and
cL = c0 (1 + 3k) = 29 400 N/m (1 + 3 0.1667) = 44 100 N/m for the loaded vehicle fol-
and
low from Eqs. (5.34) and (5.35). Now, the undamped natural frequency for the unloaded
p
p
0
0
= cL /(M + 4M ) = 1.46 Hz are
f0C
= c0 /M = 1.46 Hz and the loaded vehicle f0C
in deed the same.
of the chassis
where
zC (t)
To achieve the same damping rates for the chassis and the wheel model dierent values
for the damping parameter
dS
were needed.
With increased damping the overshoot eect in the time history of the chassis displacement and the wheel load becomes smaller and smaller till it vanishes completely at
and
W = 1.
C = 1
=1
=1
=1
the eigenvalues of the single mass oscillator change from complex to real.
Thus, producing a non oscillating solution without any sine and cosine terms.
C =1
N
dopt
S Comfort = 5292
m/s
and
W =1
N
dopt
S Safety = 6928
m/s
will avoid an overshoot eect in the time history of the chassis displacement
in the time history of the wheel load
FT (t).
(5.41)
zC (t) or in the
optimal comfort and optimal ride safety will be dierent. Hence, a simple linear damper
can either avoid overshoots in the chassis motions or in the wheel loads.
81
Vehicle Dynamics
chassis model
350 kg
dS
20000 N/m
wheel model
C [ - ]
dS [Ns/m]
W [ - ]
dS [Ns/m]
0.25
0.50
0.75
1.00
1.25
1323
2646
3969
5292
6614
0.25
0.50
0.75
1.00
1.25
1732
3464
5196
6928
8660
displacement [mm]
200
15
100
10
50
-50
-5
-100
50 kg
220000 N/m
displacement [mm]
20
150
20000 N/m
dS
-10
0
0.5
t [s]
0.05
0.1
t [s] 0.15
0.1
t [s]
acceleration [g]
1.5
6000
5000
0.5
4000
0
3000
2000
-0.5
1000
-1
0
0
0.5
t [s]
1.5
0.05
0.15
zC (t)
which surely is not a desirable conguration, because then, it takes a very long
82
z(t
= 0) = 0.
z(t = 0) = z0
z0
zS
tE
z(t)
errors allows to judge the attenuation. If the overall quadratic error becomes a minimum
t=t
Z E
=
z(t)2 dt M in ,
(5.42)
t=0
the system approaches the steady state position as fast as possible. In theory
tE
tE
To judge ride comfort and ride safety the hub motion of the chassis zC , its acceleration
zC and the variations of the dynamic wheel load FTD can be used. In the absence of road
D
irregularities zR = 0 the dynamic wheel load from Eq. (5.8) simplies to FT = cT zW .
Hence, the demands
2C
t=t
Z E
2
g1 zC
g2 zC
2 i
dt M in
(5.43)
t=0
and
t=t
Z E
2S =
cT zW
2
dt M in
(5.44)
t=0
will guarantee optimal ride comfort and optimal ride safety. By the factors
g1
and
g2
the
2
zC + 2C zC ,
zC = 0C
where, the system parameter
C 0C = dS /(2M )
M , dS
and
cS
(5.45)
0C = cS /M .
C =
Then, the
2C =
t=t
Z E
t=0
t=t
Z E
=
t=0
2
g12 0C
zC + 2C zC
zC zC
g12
2
2 2
(0C
)
g12
2
0C
i
+ g22 zC2 dt
g22
2C
g12
g12
2
0C
2C
2
(2C )
zC
zC
(5.46)
M in ,
83
Vehicle Dynamics
where
xTC = zC zC
is the state vector of the chassis model. In a similar way Eq. (5.44)
can be transformed to
2S
t=t
Z E
c2T
2
zW
t=t
Z E
dt =
zW zW
c2T 0
0 0
#"
t=0
t=0
where
"
xTW = zW zW
zW
zW
#
M in ,
(5.47)
The problems given in Eqs. (5.46) and (5.47) are called disturbance-reaction problems.
They can be written in a more general form
t=t
Z E
xT (t) Q x(t) dt M in
(5.48)
t=0
where
x(t)
Q = QT
single mass oscillators described by Eq. (5.9) the state equation reads as
0
1
z
z
=
.
2
0 2
z
z
|
{z
} | {z }
| {z }
x
x
A
(5.49)
the time response of the system exposed to the initial disturbance x(t = 0) =
x0 vanishes x(t ) = 0, and the integral in Eq.(5.48) can be solved by
For tE
t=t
Z E
(5.50)
t=0
where the symmetric matrix
R = RT
AT R + R A + Q = 0 .
(5.51)
0 02
1 2
R11 R12
R12 R22
+
R11 R12
R12 R22
0
1
02 2
+
Q11 Q12
Q12 Q22
.
(5.52)
R11 =
84
1
+
2
0 4
Q11 Q12 +
02
Q22 ,
4
(5.53)
R12 =
Q11
,
202
R22 =
Q11
Q22
+
.
2
4 0
4
(5.54)
t=t
Z E
x0 = [ z0 0 ]T
x (t) Q x(t) dt =
z02
R11 =
z02
t=0
+
2
0 4
02
Q11 Q12 +
Q22 .
4
(5.55)
Now, the integral in Eq. (5.46) evaluating the ride comfort is solved by
2C
2
2
C
1
0C
2
2
2
2
2 2
2
=
+
g1 0C + g2 g1 0C 2 C +
g (2C )
2
0C
4C
4C 1
#
"
2 !
2
g
g
2
2
0C
2
2
g12 +
+
C 0C .
= z0C
0C
2
2
4C
0C
0C
2
z0C
g1 = 1
and
g2 = 0
(5.56)
C 0C .
zC
is weighted
2
2 0C
2C z = z0C
0C
C
4C
which will become a minimum for
0C 0
(5.57)
C .
or
cS 0
C
0C 0 is
As mentioned before,
M .
However, a
large chassis mass is uneconomic and the suspension stiness is limited by the the loading
conditions. Hence, weighting the chassis accelerations only does not lead to a specic
result for the system parameter.
The combination of
2C z
C
z2
= 0C
0C
1
+ C
4C
0C or
2
d 2C |zC
z0C
1
=
+1 = 0.
d C
0C 4C2
(5.58)
(5.59)
0C
cS zC
g1 = M
and
g1 = 1
g2 = cS
and
M zC
and
or
g2 =
cS
2
= 0C
M
(5.60)
85
Vehicle Dynamics
2C
2
z0C
2
0C
0C
+ C 0C
2C
0C 0.
(5.61)
d 2C |zC
2
2
0C
= z0C
d C
C =
0C
+ 0C
2C2
will satisfy
= 0.
(5.62)
1
2
2
(5.63)
or a damping parameter of
1
opt C = 2 2
dS Comfort
2 cS M ,
(5.64)
will provide optimal comfort by minimizing the merit function in Eq. (5.61).
For the passenger car with
parameter will amount to
which is
70%
M = 350 kg
and
cS = 20 000 N/m
C = 12 2
N
dopt
S Comfort = 3742
m/s
(5.65)
2S
z2
= 0W
0W
W
1
+
4W
c2T
(5.66)
where the model parameter m, cS , dS and cT where replaced by the undamped natural
2
frequency 0W = (cS + cT )/m and by the damping ratio W = W 0W = dS /(2m).
2
A soft tire cT 0 make the safety criteria Eq. (5.66) small S 0 and thus, reduces the
dynamic wheel load variations. However, the tire spring stiness can not be reduced to
arbitrary low values, because this would cause too large tire deformations. Small wheel
masses
m 0
cS
will increase
0W
and thus,
reduce the safety criteria Eq. (5.66). The use of light metal rims improves, because of
wheel weight reduction, the ride safety of a car. Hard body suspensions contradict a good
driving comfort.
With xed values for
if
cT
and
0W
2
2S
z0W
=
W
0W
1
1+ 2
4W
c2T = 0
(5.67)
W =
86
1
2
(5.68)
p
dopt
(cS + cT ) m
S Safety =
(5.69)
will guarantee optimal ride safety by minimizing the merit function in Eq. (5.66).
For the passenger car with
M = 350 kg
and
cS = 20 000 N/m
W = 12
N
dopt
S Safety = 3464
m/s
which is
50%
(5.70)
sky
zC
zC
dB
dW
cS
zW
zR
cS
dS
FD
zW
m
cT
zR
a) Standard Damper
m
cT
Figure 5.8: Quarter car model with standard and sky hook damper
To take this situation into account the simple quarter car models of section 5.2.3 must be
combined to a more enhanced model, Fig. 5.8a.
Assuming a linear characteristics the suspension damper force is given by
FD = dS (zW zC ) ,
where
(5.71)
87
Vehicle Dynamics
The sky hook damping concept starts with two independent dampers for the body and
the wheel/axle mass, Fig. 5.8b. A practical realization in form of a controllable damper
will then provide the damping force
FD = dW zW dC zC ,
dS
(5.72)
dW
and
dC
are available.
The equations of motion for the quarter car model are given by
M zC = FS + FD M g ,
(5.73)
m zW = FT FS FD m g ,
M, m
ments, and g
where
zC , zW
FS = FS0 + cS (zW zC ) ,
where
FS0 = mC g
cS
(5.74)
FT = FT0 + cT (zR zW ) ,
where
FT0 = (M + m) g
cS
FT 0
(5.75)
zR
describes
zC
zW
z
C
zW
| {z
x
= 0
cS
cS
M
cS
cS +cT
m
m
{z
A
dMC
dC
m
zC
h i
1
zW + 0 zR ,
0
dW
z
| {z }
M C
u
dW
cT
m
zW
m
} | {z }
| {z
}
x
B
(5.76)
of the state matrix A will characterize the eigen dynamics of the quarter
car model. In case of complex eigenvalues the damped natural eigenfrequencies are given
88
Damping ratio = D
Frequencies [Hz]
12
10
Wheel
350 kg
0.8
0.7
0.6
dS 0.5
20000 N/m
50 kg
220000 N/m
Chassis
0.4
0.2
3880
3220
1000
2000
3000
4000
5000
1000
2000
3000
4000
5000
dS [N/(m/s)]
dS [N/(m/s)]
Figure 5.9: Quarter car model with standard damper
dC = dS
and
dW = dS
relative damping
standard damper described by Eq. (5.71). Fig. 5.9 shows the eigenfrequencies
and the damping ratios
= D
f = /(2)
dS .
W = 0.5
dS = 3220 N/(m/s)
3742 N/(m/s)
and
3464 N/(m/s)
Eqs. (5.65) and (5.70) with the single mass models. Hence, the very simple single mass
models can be used for a rst damper layout. Usually, as it is here, optimal ride comfort
and optimal ride safety cannot achieved both by a standard linear damper.
The sky-hook damper, modeled by Eq. (5.72), provides with
rameter. Their inuence to the eigenfrequencies
dW
and
dS
is shown in
Fig. 5.10.
The the sky-hook damping parameter
dC , dW
mainly depends on
dW .
mainly depends on
dC ,
wheel motion can be adjusted to nearly each design goal. Here, a sky-hook damper with
dC = 3900 N/(m/s) and dW = 3200 N/(m/s) would generate the damping ratios dC = 0.7
and dW = 0.5 hence, combining ride comfort and ride safety within one damper layout.
89
Vehicle Dynamics
Damping ratios C, W
Frequencies [Hz]
12
10
0.8
0.7
350 kg
dC
50 kg
220000 N/m
2
0
1000
2000
3000
4000
4500
4000
3500
3000
2500
2000
1500
dC 0.6
dW 0.5
20000 N/m
dC [N/(m/s)]
5000
0.4
0.2
0
W
0
1000
2000
3000
4000
5000
dW [N/(m/s)]
dW [N/(m/s)]
FD = dW zW dC zC + =
dW zW dC zC
(zW zC ) = dS (zW zC )
zW zC
|
{z
}
dS
(5.77)
the sky-hook damper can be realized by a standard damper in the form of Eq. (5.71). The
new damping parameter dS now nonlinearly depends on the absolute vertical velocities of
the chassis and the wheel dS = dS (zC , zW ). As, a standard damper operates in a dissipative
mode only the damping parameter will be restricted to positive values, dS > 0. Hence, the
passive realization of a sky-hook damper will only match with some properties of the ideal
damper law in Eq. (5.72). But, compared with the standard damper it still can provide a
better ride comfort combined with an increased ride safety.
M zC = FS + FD M g
m zW = FT FS FD m g ,
g = 9.81 m/s2 labels the constant of gravity, M , m are the masses of the chassis
the wheel, FS , FD , FT describe the spring, the damper, and the vertical tire force,
where
and
90
(5.78)
zC
nonlinear spring
nonlinear damper
FS
FD
FS
FD
zW
zR
cT
Figure 5.11: Quarter car model with nonlinear spring and damper characteristics
zC
zW
equilibrium position.
In extension to Eq. (5.32) the spring characteristics is modeled by
FS =
FS0
c0 u
c0 u
where
FS0 = M g
1 + kr
1 + kc
u
4ur
2 !
u
4uc
2 !
u<0
(5.79)
u0
u = zW zC
(5.80)
u < 0 marks tension (rebound), and u 0 compression. Two sets of kr , ur and kc , uc dene the spring nonlinearity during rebound and
compression. For kr = 0 and kc = 0 a linear spring characteristics is obtained.
FD (v) =
where
d0
d0 v
1 pr v
v <0,
d0 v
1 + pc v
v 0,
v = 0,
(5.81)
v = zW zC .
(5.82)
The sign convention of the damper velocity was chosen consistent to the spring travel.
Hence, rebound is characterized by
v<0
and compression by
v 0.
The parameter
pr
91
Vehicle Dynamics
and
pc
make it possible to model the damper nonlinearity dierently in the rebound and
pr = 0
and
pc = 0.
The nonlinear spring design in Section 5.3.2 holds for the compression mode. Hence,
kc = k = 0.1667.
By setting
a) Spring
b) Damper
5000
cS = 34300 N/m
6000 c0 = 29400 N/m
ur = 0.05 m
5000 kr = 0
uc = 0.05 m
4000 kc = 0.1667
FD [N/m]
FS [N/m]
7000
2500
d0 = 4200 N/(m/s)
pr = 0.4 1/(m/s)
pc = 1.2 1/(m/s)
3000
2000
-2500
1000
0
-0.1
rebound
u<0
-0.05
compression
u>0
0.05
u [m]
0.1
-5000
-1
rebound
v<0
-0.5
compression
v>0
0.5
v [m/s]
(dS )C
opt =
(dS )Sopt
2 cS M =
M = 350 kg , m = 50 kg
and
p
(cS + cT ) m = (18 000 + 220 000) 50 = 3570 N/(m/s) .
1
The mean value d0 = 4200N/(m/s) may serve as compromise. With pr = 0.4(m/s)
and
pc = 1.2 (m/s)1 the nonlinearity becomes more intensive in compression than rebound,
Fig. 5.12b.
5.5.2 Results
The quarter car model is driven with constant velocity over a single obstacle. Here, a
cosine shaped bump with a height of
H = 0.08 m
and a length of
L = 2.0 m
was used.
92
7.1
6.0
6660
6160
6000
0.02
5000
0
4000
-5
3000
-0.02
2000
-10
-15
0.5
1.5
-0.04
linear
nonlinear
1000
0
0.5
time [s]
1.5
-0.06
0.5
time [s]
instead of
1.5
time [s]
v = 20 km h
6660 N ).
t 0.25 s
can be avoided. While crossing the bump large damper velocities occur. Here, the degressive damper characteristics provides less damping compared to the linear damper which
increases the suspension travel.
7000
0.04
6000
0.02
5000
0
4000
-5
3000
0
-0.02
-0.04
2000
-10
-15
linear,
low damping
1000
0
0.5
time [s]
1.5
-0.06
nonlinear
0.5
time [s]
1.5
-0.08
0.5
1.5
time [s]
the peaks in the chassis acceleration and in the wheel load, but then the attenuation of
the disturbances will take more time. Fig. 5.14. Which surely is not optimal.
93
6 Longitudinal Dynamics
6.1 Dynamic Wheel Loads
6.1.1 Simple Vehicle Model
The vehicle is considered as one rigid body which moves along an ideally even and horizontal road. At each axle the forces in the wheel contact points are combined in one
normal and one longitudinal force.
S
h
Fz1
Fx1
mg
Fx2
a2
a1
Fz2
If aerodynamic forces (drag, positive and negative lift) are neglected at rst, the equations
of motions in the
where
x-, z -plane
will read as
m v = Fx1 + Fx2 ,
(6.1)
0 = Fz1 + Fz2 m g ,
(6.2)
(6.3)
a1 + a2
is the
These are only three equations for the four unknown forces
But, if we
insert Eq. (6.1) in Eq. (6.3), we can eliminate two unknowns at a stroke
0 = Fz1 a1 Fz2 a2 + m v h .
94
(6.4)
The equations Eqs. (6.2) and (6.4) can be resolved for the axle loads now
Fz1 = m g
h
a2
m v ,
a1 + a2
a1 + a2
(6.5)
Fz2 = m g
h
a1
+
m v .
a1 + a2
a1 + a2
(6.6)
st
Fz1
= mg
a2
,
a1 + a2
st
Fz2
= mg
a1
a1 + a2
(6.7)
describe the weight distribution according to the horizontal position of the center of
gravity. The height of the center of gravity only inuences the dynamic part of the axle
loads,
dyn
Fz1
= m g
When accelerating
v > 0,
h
v
,
a1 + a2 g
dyn
Fz2
= +m g
h
v
.
a1 + a2 g
(6.8)
v < 0.
v
x
Fx1
mg
Fz1
a1
a2
h
Fx2
Fz2
For a vehicle on a grade, Fig.6.2, the equations of motion Eq. (6.1) to Eq. (6.3) can easily
be extended to
(6.9)
95
Vehicle Dynamics
where
denotes the grade angle. Now, the axle loads are given by
Fz1 = m g cos
h
a2 h tan
m v ,
a1 + a2
a1 + a2
(6.10)
Fz2 = m g cos
a1 + h tan
h
+
m v ,
a1 + a2
a1 + a2
(6.11)
where the dynamic parts remain unchanged, whereas now the static parts also depend on
the grade angle and the height of the center of gravity.
FD1
FD2
FAR
h
mg
Fx1
a2
a1
Fz1
Fx2
Fz2
v > 0,
(6.12)
FAR
and
FD1 , FD2
describe the air resistance and the down forces. For the dynamic
Fz1 = FD1 + m g
a2
h
(m v + FAR ) ,
a1 + a2
a1 + a2
(6.13)
Fz2 = FD2 + m g
a1
h
+
(m v + FAR ) .
a1 + a2
a1 + a2
(6.14)
FD1 , FD2
96
FAR
Fz1 0
and
Fz2 0
v
a2
cos sin
g
h
and
v
a1
cos sin .
g
h
(6.15)
a1
cos
h
v
+ sin
g
and the
> 0)
position a1 , a2 , h of
a2
cos .
h
(6.16)
and decelerations (v
< 0) are
For v
0
a2
a1
tan
h
h
(6.17)
|Fx1 | Fz1
where the same friction coecient
and
|Fx2 | Fz2
(6.18)
case
Fx1 = Fz1
and
Fx2 = Fz2
(6.19)
(6.20)
v
=
g max
That means climbing (v
a friction coecient
> 0, > 0)
tan ||.
cos sin .
or downhill stopping (v
< 0, < 0)
(6.21)
requires at least
97
Vehicle Dynamics
According to the vehicle dimensions and the friction values the maximal acceleration or
deceleration is restricted either by Eq. (6.16) or by Eq. (6.21).
If we take aerodynamic forces into account, the maximum acceleration and deceleration
on a horizontal road will be limited by
FD1 FD2
+
1 +
mg
mg
FAR
mg
v
g
FD1 FD2
1 +
+
mg
mg
FAR
.
mg
(6.22)
In particular the aerodynamic forces enhance the braking performance of the vehicle.
Fx1 = 0
and
Fx2 = Fz2
m v R WD = m g
a1
h
v R WD
+
a1 + a2 a1 + a2 g
,
(6.23)
where the subscript R WD indicates the rear wheel drive. Hence, the maximum acceleration
for a rear wheel driven vehicle is given by
v R WD
=
g
By setting
Fx1 = Fz1
and
Fx2 = 0,
h
1
a1 + a2
a1
.
a1 + a2
(6.24)
v F WD
=
g
a2
,
h
a1 + a2
1+
a1 + a2
where the subscript F WD denotes front wheel drive. Depending on the parameter
a2 and h the accelerations may be limited by the tilting condition vg ah2 .
(6.25)
, a1 ,
The maximum accelerations of a single axle driven vehicle are plotted in Fig. 6.4. For rear
98
v/g
0.8
FWD
RWD
FWD
0.6
0.4
RWD
0.2
0
0.2
0.4
0.6
0.8
a2 / (a1+a2)
Fx1 = Fz1
and
Fx2 = 0
m v F WB = m g
h
v F WB
a2
a1 + a2 a1 + a2 g
,
(6.26)
where the subscript F WB indicates front wheel braking. Then, the maximum deceleration
is given by
v F WB
=
g
If only the rear axle is braked (Fx1
deceleration
h
1
a1 + a2
a2
.
a1 + a2
= 0, Fx2 = Fz2 ),
v R WB
=
g
h
1+
a1 + a2
(6.27)
a1
,
a1 + a2
(6.28)
where the subscript R WB denotes a braked rear axle. Depending on the parameters
v
and h, the decelerations may be limited by the tilting condition
ah1 .
g
, a1 ,
a2 ,
The maximum decelerations of a single axle braked vehicle are plotted in Fig. 6.5. For
passenger cars the load distribution parameter
to
0.6.
will be achieved. This is a quite large value compared to the deceleration range of a braked
rear axle which is in the range of
v/g
= 0.49
to
v/g
= 0.33.
99
Vehicle Dynamics
range of
load
distribution
v/g
-0.2
FWB
-0.4
-0.6
-0.8
-1
RWB
0
0.2
0.4
0.6
0.8
a2 / (a1+a2)
Fx1
Fx2
v
=
+
.
g
mg
mg
(6.29)
Fx1
and
Fx2 .
Fx1
and
Fx2
Fx1 = Fz1
and
Fx2 = Fz2 .
(6.30)
and
Fx1
=
mg
Fx2
=
mg
a2
v
h
g
a1
v
+
h
g
h
a1 + a2
(6.31)
h
.
a1 + a2
(6.32)
With Eq. (6.31) and Eq. (6.32) one gets from Eq. (6.29)
v
= ,
g
where it has been assumed that
Fx1
and
Fx2
(6.33)
Fx1
v
=
mg
g
100
a2
v
h
g
h
a1 + a2
(6.34)
and
Fx2
v
=
mg
g
a1
v
+
h
g
h
.
a1 + a2
(6.35)
B2/mg
that grant the same skid resistance at both axles can be calculated now.
braking
Fx1/mg
-a1/h
dFx2
dFx1 0
-1
-2
B1/mg
a =1.15
1
driving
a =1.35
2
h=0.55
=1.20
tilting limits
Fx2/mg
a2/h
v/g
= a1 /h and v/g
= +a2 /h, no longitudinal forces can be applied
at the lifting axle. The initial gradient only depends on the steady state distribution of
the wheel loads. From Eqs. (6.34) and (6.35) it follows
Fx1
a2
v
h
mg
=
2
v
h
g a1 + a2
d
g
(6.36)
Fx2
a1
v
h
mg
=
+2
.
v
h
g a1 + a2
d
g
(6.37)
and
101
Vehicle Dynamics
For
v/g
=0
d Fx2
a1
.
=
d Fx1 0
a2
(6.38)
limitation
Fx2/mg
Fx2/mg
constant
distribution
Fx1/mg
Fx2/mg
Fx1/mg
Fx1/mg
reduction
6.3.5 Anti-Lock-Systems
On hard braking maneuvers large longitudinal slip values occur. Then, the stability and/or
steerability is no longer given because nearly no lateral forces can be generated. By controlling the brake torque or brake pressure respectively, the longitudinal slip can be restricted
to values that allow considerable lateral forces.
Here, the angular wheel acceleration
of the wheel are derived from the measured angular speeds of the wheel by dierentiation.
The rolling condition is fullled with a longitudinal slip of
rD = x
102
sL = 0.
Then
(6.39)
holds, where
rD
|
x| = g .
a simple
||
1
|
x| .
rD
(6.40)
Because no reliable possibility to determine the local friction coecient between tire and
road has been found until today, useful information can only be gained from Eq. (6.40)
at optimal conditions on dry road. Therefore, the longitudinal slip is used as a second
control variable.
In order to calculate longitudinal slips, a reference speed is estimated from all measured
wheel speeds which is used for the calculation of slip at all wheels, then. This method
is too imprecise at low speeds. Therefore, no control is applied below a limit velocity.
Problems also arise when all wheels lock simultaneously for example which may happen
on icy roads.
The control of the brake torque is done via the brake pressure which can be
held,
or
decreased
increased,
at the rear axle. Here, the break pressure at both wheels is controlled by the wheel running
on lower friction. Thus, at least the brake forces at the rear axle cause no yaw torque.
However, the maximum achievable deceleration is reduced by this.
z1
and
z2
xA ,
vertical motion
zA
and pitch
A .
The coordinates
describe the vertical motions of wheel and axle bodies relative to the body.
The longitudinal and rotational motions of the wheel bodies relative to the body can be
described via suspension kinematics as functions of the vertical wheel motion:
x1 = x1 (z1 ) , 1 = 1 (z1 ) ;
x2 = x2 (z2 ) , 2 = 2 (z2 ) .
The rotation angles
R1
and
R2
The forces between wheel body and vehicle body are labeled
drive torques
(6.41)
FF 1
and
FF 2 .
At the wheels
MA1 , MA2 and brake torques MB1 , MB2 , longitudinal forces Fx1 , Fx2 and
Fz1 , Fz2 apply. The brake torques are directly supported by the wheel
bodies, whereas the drive torques are transmitted by the drive shafts to the vehicle body.
103
Vehicle Dynamics
zA
xA
FF1
MA1
z1
MB1
MA1
hR
R1
MB1
Fz1
z2
MB2
MA2
Fx1
R
a1
FF2
MA2
R2
a2
MB2
Fz2
Fx2
The forces and torques that apply to the single bodies are listed in the last column of the
tables 6.1 and 6.2.
The velocity of the vehicle body and its angular velocity are given by
v0A,0
x A
0
= 0 + 0 ;
0
zA
0
= A .
0
0A,0
(6.42)
At small rotational motions of the body one gets for the velocities of the wheel bodies
and wheels
x1
z1
v0RK1 ,0 = v0R1 ,0
x A
0
hR A
+
0
= 0 + 0 +
0
zA
a1 A
x2
z2
v0RK2 ,0 = v0R2 ,0
x A
0
hR A
+
0
= 0 + 0 +
0
zA
+a2 A
z1
0
z1
z2
0
z2
(6.43)
(6.44)
The angular velocities of the wheel bodies and wheels are obtained from
0RK1 ,0
0
0
= A + 1
0
0
and
0R1 ,0
0
0
0
= A + 1 + R1
0
0
0
(6.45)
104
as well as
0
0
= A + 2
0
0
0RK2 ,0
0R2 ,0
and
0
0
0
= A + 2 + R2
0
0
0
(6.46)
z =
x A zA A 1 R1 2 R2
T
(6.47)
the velocities and angular velocities given by Eqs. (6.42), (6.43), (6.44), (6.45), and (6.46)
can be written as
v0i =
7
X
v0i
j=1
zj
zj
and
0i =
7
X
0i
j=1
zj
zj
(6.48)
0i
v0i
and partial angular velocities
for the ve bodies
zj
zj
j = 1(1)7
partial velocities
bodies
chassis
mA
wheel body
front
mRK1
wheel
front
mR1
wheel body
rear
mRK2
wheel
rear
mR2
x A
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
zA
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
A
0
0
0
hR
0
a1
hR
0
a1
hR
0
a2
hR
0
a2
z1
0
0
0
x1
z1
0
1
x1
z1
0
1
0
0
0
0
0
0
i = 1(1)5
v0i /zj
R1
z2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
x2
0
z2
0
0
0
1
x2
0
z2
0
0
0
1
applied forces
R2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Fie
0
0
FF 1 +FF 2 mA g
0
0
FF 1 mRK1 g
Fx1
0
Fz1 mR1 g
0
0
FF 2 mRK2 g
Fx2
0
Fz2 mR2 g
Q can
be calculated.
M (i, j) =
T
5
X
v0k
k=1
zi
T
5
X
v0k
0k
0k
mk
+
k
;
zj
zj
zi
k=1
i, j = 1(1)7 ;
(6.49)
105
Vehicle Dynamics
wheel body
front
RK1
wheel
front
R1
wheel body
rear
RK2
wheel
rear
R2
x A
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
zA
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
z1
0
0
0
0
R1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
z1
0
0
1
z1
0
0
0
0
0
0
0
0i /zj
applied torques
z2 R2
Mie
0
0
0
0
0 MA1MA2 a1 FF 1 +a2 FF 2
0
0
0
0
0
0
0
0
MB1
0
0
0
0
0
0
0
0
MA1 MB1 R Fx1
0
0
0
0
0
0
2
0
M
B2
z2
0
0
0
0
0
0
2
1
M
M
A2
B2 R Fx2
z2
0
0
0
Q(i) =
T
5
X
v0k
k=1
zi
Fke
T
5
X
0k
k=1
zi
Mke ;
i = 1(1)7 .
(6.50)
Then, the equations of motion for the plane vehicle model are given by
M z = Q .
(6.51)
6.4.3 Equilibrium
With the abbreviations
m1 = mRK1 + mR1 ;
m2 = mRK2 + mR2 ;
mG = mA + m1 + m2
(6.52)
and
h = hR + R
(6.53)
(6.54)
x1
z1
Fx1 m1 g +
106
1
(MA1
z1
R Fx1 ) ;
(6.55)
Q(6) = Fz2 FF 2 +
x2
z2
Fx2 m2 g +
2
(MA2
z2
R Fx2 ) ;
(6.56)
MA1 = 0 ;
MA2 = 0 ;
MB1 = 0 ;
MB2 = 0
(6.57)
from Eqs. (6.54), (6.55) and (6.56) one gets the steady state longitudinal forces, the spring
preloads, and the wheel loads
0
Fx1
= 0;
FF0 1 =
0
Fz1
0
Fx2
= 0;
b
a+b
= m1 g +
a
a+b
FF0 2 =
mA g ;
a2
a1 +a2
0
Fz2
mA g ;
mA g ;
= m2 g +
(6.58)
a1
a1 +a2
mA g .
R R1 = x A hR A +
R R2 = x A hR A +
x1
z1
x2
z2
z1 ;
(6.59)
z2
hold. In steady state the pitch motion of the body and the vertical motion of the wheels
reach constant values
A = Ast = const. ,
z1 = z1st = const. ,
z2 = z2st = const.
(6.60)
R R1 = x A ;
R R2 = x A .
(6.61)
With Eqs. (6.60), (6.61) and (6.53) the equation of motion (6.51) results in
a
a
mG xA = Fx1
+ Fx2
;
a
a
0 = Fz1
+ Fz2
;
hR (m1 +m2 ) xA + R1
x
A
R
+ R2
x
A
R
a
a
a
a
= a Fz1
+ b Fz2
(hR + R)(Fx1
+ Fx2
);
(6.62)
x1
z1
x2
z2
m1 xA +
m2 xA +
1
z1
2
z2
R1 xRA
a
Fz1
FFa 1
x1
z1
a
Fx1
R1
x
A
R
a
= MA1 MB1 R Fx1
;
R2
x
A
R
a
= Fz2
FFa 2 +
R2
x
A
R
a
= MA2 MB2 R Fx2
;
x2
z2
a
Fx2
+
1
(MA1
z1
2
(MA2
z2
a
R Fx2
);
a
R Fx1
)
(6.63)
(6.64)
107
Vehicle Dynamics
where the steady state spring forces, longitudinal forces, and wheel loads have been separated into initial and acceleration-dependent terms
a
0
st
;
+ Fxi
= Fxi
Fxi
i = 1, 2 .
(6.65)
a
With given torques of drive and brake the vehicle acceleration x
A , the wheel forces Fx1
,
a
a
a
a
a
Fx2 , Fz1 , Fz2 and the spring forces FF 1 , FF 2 can be calculated from Eq. (6.62), Eq. (6.63)
and Eq. (6.64)
Via the spring characteristics which have been assumed as linear the accelerationdependent forces also cause a vertical displacement and pitch motion of the body besides
the vertical motions of the wheels,
FFa 1
FFa 2
a
Fz1
a
Fz2
Especially the pitch of the vehicle
= cA1 z1a ,
= cA2 z2a ,
= cR1 (zAa a Aa + z1a ) ,
= cR2 (zAa + b Aa + z2a ) .
(6.66)
if too distinct.
By an axle kinematics with 'anti dive' and/or 'anti squat' properties,
brake pitch angle can be reduced by rotating the wheel body and moving the wheel center
in longitudinal direction during the suspension travel.
x-, z -plane,
Fig. 6.9.
pitch pole
108
7 Lateral Dynamics
7.1 Kinematic Approach
7.1.1 Kinematic Tire Model
When a vehicle drives through a curve at low lateral acceleration, small lateral forces
will be needed for course holding. Then, hardly lateral slip occurs at the wheels. In the
ideal case at vanishing lateral slip the wheels only move in circumferential direction. The
velocity component of the contact point in the lateral direction of the tire vanishes then
vy = eTy v0P = 0 .
(7.1)
This constraint equation can be used as 'kinematic tire model' for course calculation of
vehicles moving in the low lateral acceleration range.
M,
Fig. 7.1.
At slowly moving vehicles the lay out of the steering linkage is usually done according to
the Ackermann geometry. Then, the following relations apply
a
a
and
tan 2 =
,
(7.2)
R
R+s
where s labels the track width and a denotes the wheel base. Eliminating the curve radius
R, we get
a
a tan 1
tan 2 =
or
tan 2 =
.
(7.3)
a
a + s tan 1
+s
tan 1
a
A
a
The deviations 42 = 2 2 of the actual steering angle 2 from the Ackermann steering
A
angle 2 , which follows from Eq. (7.3), are used, especially on commercial vehicles, to
tan 1 =
M,
axis of the vehicle is called side slip angle. The side slip angle at point
tan P =
where
x
R
P to
or
tan P =
x
tan 1 ,
a
is given by
(7.4)
109
Vehicle Dynamics
v
P
P
a
x
Rmax ,
whereas a point on the inner side of the vehicle at the location of the rear
Rmin .
Hence, it holds
2
Rmax
= (Rmin + b)2 + (a + f )2 ,
where
a, b are the wheel base and the width of the vehicle, and f
(7.5)
species the distance from
the front of the vehicle to the front axle. Then, the space requirement
can be specied as a function of the cornering radius
4R = Rmax Rmin =
The space requirement
4R
Rmin
110
2.5
(7.6)
4R = Rmax Rmin
ma
Rmin
R [m]
5
4
3
2
1
0
10
20
30
R min [m]
40
50
111
Vehicle Dynamics
x0 , y0 , z0
= y
R
r02,0
and the rotation matrix
(7.7)
A02
Here, the tire radius
cos sin 0
cos 0 .
= sin
0
0
1
(7.8)
x, y
a
b
A1
y2
x2
A2
y0
y3
A3
x0
112
with
r21,2
a
= 0
0
(7.9)
with
cos sin 0
cos 0
= sin
0
0
1
A21
describe the position and the orientation of the front axle, where
wheel base and
a = const
(7.10)
labels the
r2K,2
b
= 0
0
(7.11)
c
= 0
0
(7.12)
cos sin 0
cos 0
= sin
0
0
1
(7.13)
rK3,2
and
with
A23
dene the position and the orientation of the trailer axis, with
between vehicle and trailer, and
coupling point
b, c
to the
3.
v02,2
vx2
= 0 .
0
(7.14)
v02,0 = r02,0
x
= y .
0
(7.15)
vx2
cos vx2
0 = sin vx2
0
0
(7.16)
113
Vehicle Dynamics
compared to Eq. (7.15) results in two rst order dierential equations for the position
coordinates
and
y
x = vx2 cos ,
y = vx2 sin .
(7.17)
vx2
0
a
vx2
= 0 + 0 0 = a .
0
0
0
| {z }
| {z } | {z }
v02,2
02,2
r21,2
(7.18)
v01,2
(7.19)
ex1,2
cos
= sin
0
and
ey1,2
sin
= cos
0
(7.20)
dene the longitudinal and lateral direction at the front axle. According to Eq. (7.1) the
velocity component lateral to the wheel must vanish,
(7.21)
(7.22)
remains. From Eq. (7.21) a rst order dierential equation follows for the yaw angle
vx2
tan .
a
(7.23)
k = 1/R
tan =
a
= ak .
R
(7.24)
= vx2 k .
With the curvature gradient
k = k(t) = kC
114
(7.25)
t
,
T
(7.26)
the entering of a curve is described as a continuous transition from a straight line with
k=0
the curvature
k = kC .
The yaw angle of the vehicle can be calculated by simple integration now
vx2 kC t2
,
T
2
angle, (t = 0) = 0,
(t) =
where at time
t=0
a vanishing yaw
(7.27)
has been assumed. Then, the
position of the vehicle follows with Eq. (7.27) from the dierential equations Eq. (7.17)
Zt=T
x = vx2
cos
vx2 kC t2
T
2
Zt=T
dt ,
y = vx2
t=0
sin
vx2 kC t2
T
2
dt .
(7.28)
t=0
From Eq. (7.24) the necessary steering angle can be calculated, too. If only small steering
angles are necessary for driving through the curve, the
= (t) a k = a kC
t
T
(7.29)
holds, i.e. the driving through a clothoide is manageable by a continuous steer motion.
r02,0 = v02,0
02,0
(7.30)
Eqs. (7.16) and (7.19). The position vector from the rear axle to the axle of the trailer is
given by
r2K,2
and
rK3,3
b c cos
c sin ,
=
0
(7.31)
are dened in Eq. (7.12). The time derivative of Eq. (7.31) results
in
c cos
c sin
0
= 0 c sin = c cos .
0
0
{z
}
| {z } |
23,2
A23 rK3,3
r23,2
(7.32)
x2 , y2 , z2 now
vx2
0
b c cos
c sin
vx2 + c sin (+
)
0
0
0
{z
} |
{z
}
| {z } | {z } |
v02,2
02,2
r23,2
r23,2
v03,2
(7.33)
115
Vehicle Dynamics
The longitudinal and lateral direction at the trailer axle are dened by the unit vectors
ex3,2
cos
= sin
0
ey3,2
and
sin
= cos .
0
(7.34)
)
+ cos b c cos (+
)
(7.35)
)
(7.36)
remains. If Eq. (7.23) is inserted into Eq. (7.35) now, one will get a rst order dierential
equation for the bend angle
vx2
=
a
a
sin +
c
b
cos + 1 tan
c
.
(7.37)
The dierential equations Eq. (7.17) and Eq. (7.23) describe position and orientation
x0 , y0
within the
plane. The position of the trailer relative to the vehicle follows from
Eq. (7.37).
front axle
rear axle
trailer axle
20
[m]
10
0
-30
-20
-10
10
20
30 [m] 40
50
60
30
[o]
20
10
0
10
15
20
[s]
25
30
a, b , c ,
= (t),
can be calculated by numerical integration of the dierential equations Eqs. (7.17), (7.23)
and (7.37). If the steering angle is slowly increased at constant driving speed, the vehicle
drives a gure which will be similar to a clothoide, Fig. 7.5.
116
B,
described by
v0C,B
where
v cos
= v sin
0
and
0F,B
0
= 0 ,
denotes the side slip angle of the vehicle measured at the center of gravity. The
=
where
(7.38)
v
,
R
(7.39)
Fx2
a2
Fy2
a1
xB
Fx1
R
yB
Fy1
In the body xed reference frame, linear and angular momentum result in
v2
m sin
= Fx1 cos Fy1 sin + Fx2 ,
R
2
v
m
cos
= Fx1 sin + Fy1 cos + Fy2 ,
R
0 = a1 (Fx1 sin + Fy1 cos ) a2 Fy2 ,
(7.40)
(7.41)
(7.42)
117
Vehicle Dynamics
where
longitudinal and vertical direction applied at the front and rear axle, and
species the
Fx1 = k FD
where
FD
k=0
and
Fx2 = (1 k) FD ,
0<k<1
k=1
(7.43)
can be modeled:
Fx1 = 0, Fx2 = FD
Fx1
k
=
Fx2
1k
Fx1 = FD , Fx2 = 0
k cos + (1k) FD
k sin FD +
sin Fy1
cos Fy1 +
Fy2
mv 2
=
sin ,
R
mv 2
=
cos ,
R
(7.44)
FD
a2
cos sin sin cos
mv 2
a1 + a2
=
.
k + (1 k) cos
R
(7.45)
a2
cos sin = sin cos
a1 + a2
or
a2
tan = tan
a1 + a2
(7.46)
holds. This fully corresponds with the Ackermann geometry. But, the Ackermann geometry applies only for small lateral accelerations. In real driving situations, the side slip
angle of a vehicle at the center of gravity is always smaller than the Ackermann side slip
a2
angle. Then, due to tan <
tan a driving force FD > 0 is needed to overcome the
a1 +a2
'cornering resistance' of the vehicle.
118
2 1
m ay
h2
mg
h1
F yL
F yR
FzL
s/2
s/2
FzR
The balance of torques at the height of the track plane applied at the already inclined
vehicle results in
s
= m ay (h1 + h2 ) + m g [(h1 + h2 )1 + h2 2 ] ,
2
where ay describes the lateral acceleration, m is the sprung mass, and small
the axle and the body were assumed, 1 1, 2 1.
(FzL FzR )
(7.47)
roll angles of
T
FzR
= 0,
(7.48)
T
FzL
= mg .
(7.49)
Using Eqs. (7.48) and (7.49) one gets from Eq. (7.47)
s
aTy
h2
2
=
1T
2T .
g
h1 + h2
h1 + h2
(7.50)
T
rises above the limit ay . Roll
T
T
of axle and body reduce the overturning limit. The angles 1 and 2 can be calculated
from the tire stiness cR and the roll stiness of the axle suspension.
The vehicle will turn over, when the lateral acceleration
ay
119
Vehicle Dynamics
If the vehicle drives straight ahead, the weight of the vehicle will be equally distributed
to both sides
stat
stat
FzR
= FzL
=
1
mg .
2
(7.51)
With
T
stat
FzL
= FzL
+ 4Fz
(7.52)
and Eqs. (7.49), (7.51), one obtains for the increase of the wheel load at the overturning
limit
4Fz =
1
mg .
2
(7.53)
4Fz = cR 4r ,
where
cR
(7.54)
Because the right tire simultaneously rebounds with the same amount, for the roll angle
of the axle
2 4r = s 1T
or
1T =
mg
2 4r
=
s
s cR
(7.55)
holds. In analogy to Eq. (7.47) the balance of torques at the body applied at the roll
center of the body yields
cW 2 = m ay h2 + m g h2 (1 + 2 ) ,
(7.56)
cW names the roll stiness of the body suspension. In particular, at the overturning
ay = aTy
aTy
mgh2
mgh2
T
2 =
+
1T
(7.57)
g cW mgh2
cW mgh2
where
limit
T
applies. Not allowing the vehicle to overturn already at ay = 0 demands a minimum of
min
roll stiness cW > cW
= mgh2 . With Eqs. (7.55) and (7.57) the overturning condition
Eq. (7.50) reads as
(h1 + h2 )
aTy
aTy
s
1
1
1
1
h2
,
=
(h1 + h2 ) h2
g
2
cR
g cW 1
cW 1 cR
(7.58)
cR
cR = m g
s
and
cW =
cW
m g h2
(7.59)
aTy
=
g
120
s
2
h2
h1 + h2 +
cW 1
1
cR
(7.60)
overturning limit ay
0.6
20
0.5
15
0.4
0.3
10
0.2
5
0.1
0
0
10
20
normalized roll stiffness c W
*
0
10
20
normalized roll stiffness c W
*
Figure 7.8: Tilting limit for a typical truck at steady state cornering
remains.
in Fig. 7.8. Even with a rigid body suspension cW , the vehicle turns over at a lateral
acceleration of ay 0.5 g . Then, the roll angle of the vehicle solely results from the tire
of one tire is
deection.
At a normalized roll stiness of
cW = 5,
ay 0.45 g
and
so reaches already 90% of the maximum. The vehicle will turn over at a roll angle of
= 1 + 2 10 then.
ay ,
centrifugal forces
will be applied to the single masses. At the simple roll model in Fig. 7.9, these are the
forces
m A ay
and
m R ay ,
where
mA
m A ay
mR
z1 = z2 .
At steady state cornering, the vehicle forces are balanced. With the principle of virtual
work
W = 0 ,
(7.61)
FF 1 , FF 2
cA
and
121
Vehicle Dynamics
b/2
b/2
zA
mA a y
A
SA
yA
FF1
FF2
h0
z2
z1
mR a y
mR a y
S2
r0
y2
Q2
S1
Q1
F y2
Fy2
F z1
y1
F y1
cQ , cR .
The work
potential
V.
W = V
via the
W = mA ay yA
mR ay (yA + hR A + y1 )2 mR ay (yA + hR A + y2 )2
21 cA z12
1
2
cA z22
(7.62)
21 cS (z1 z2 )2
21 cQ (yA + h0 A + y1 + r0 1 )2 12 cQ (yA + h0 A + y2 + r0 2 )2
2
2
21 cR zA + 2b A + z1 21 cR zA 2b A + z2 ,
where the abbreviation
hR = h0 r0
cS
of the anti roll bar, converted to the vertical displacement of the wheel centers.
The kinematics of the wheel suspension are symmetrical. With the linear approaches
y1 =
the work
y
z1 ,
z
1 =
1
z
and
y2 =
y
z2 ,
z
2 =
2
z
(7.63)
y = [ yA , zA , A , z1 , z2 ]T .
(7.64)
W = W (y)
(7.65)
Due to
W =
122
W
y = 0 .
y
(7.66)
Because of
y 6= 0,
Ky = b
results from Eq. (7.66). The matrix
2 cQ
K = 2 cQ h0
yQ c
z Q
Q
yz cQ
Q
yz cQ
cR
Q
b
c +h0 yz cQ
2 R
cA + cS + cR
2b cR h0 yz cQ
and
are given by
y Q
c
z Q
Q
b
c +h0 yz cQ
2 R
cR
2 cQ h0
2 cR
cR
(7.67)
b =
cS
mA + 2 mR
0
(m1 + m2 ) hR
mR y/z
mR y/z
cR
Q
y
b
2 cR h0 z cQ
cS
cA + cS + cR
(7.68)
ay .
(7.69)
y
y Q
=
+ r0
,
z
z
z
cA
= cA + cQ
y
z
2
,
c =
2 cQ h20
2
b
+ 2 cR
.
2
(7.70)
The system of linear equations Eq. (7.67) can be solved numerically, e.g. with MATLAB.
Thus, the inuence of axle suspension and axle kinematics on the roll behavior of the
vehicle can be investigated.
a)
b)
roll center
1 0
roll center
2 0
Figure 7.10: Roll behavior at cornering: a) without and b) with camber compensation
If the wheels only move vertically to the body at jounce and rebound, at fast cornering
the wheels will be no longer perpendicular to the track Fig. 7.10 a. The camber angles
1 > 0
and
2 > 0
which leads to a reduction of the maximally transmittable lateral forces. Thus, at more
123
Vehicle Dynamics
sportive vehicles axle kinematics are employed, where the wheels are rotated around the
the wheels around the longitudinal axis on jounce and rebound, the wheel contact points
are moved outwards, i.e against the lateral force. By this, a 'roll support' is achieved that
reduces the body roll.
roll axis
roll center front
Q1
and
Q2 ,
Fig. 7.10. The line through the roll center at the front and rear axle is called
+TT
PR0+P
PF0+P
PF0-P
PR0+PR
PR0-P
PF0+PF
PR0-PR
PF0-PF
124
On most passenger cars the chassis is rather sti. Hence, front and rear part of the chassis
are forced by an internal torque to an overall chassis roll angle. This torque aects the
wheel loads and generates dierent wheel load dierences at the front and rear axle. Due
to the degressive inuence of the wheel load to longitudinal and lateral tire forces the
steering tendency of a vehicle can be aected.
a2
a1
y0
Fy2
x2
xB
Fy1
y1
y2
yB
x1
0,
Fig. 7.13. The tire forces at the wheels of one axle are combined to one resulting
force. Tire torques, rolling resistance, and aerodynamic forces and torques, applied at the
vehicle, are not taken into consideration.
7.3.2 Kinematics
The vehicle velocity at the center of gravity can be expressed easily in the body xed
frame
xB , yB , zB
vC,B
where
v cos
= v sin ,
0
(7.71)
125
Vehicle Dynamics
The velocity vectors and the unit vectors in longitudinal and lateral direction of the axles
are needed for the computation of the lateral slips. One gets
ex1 ,B
cos
= sin ,
0
and
ey1 ,B
sin
= cos ,
0
ex2 ,B
where
and
a1
and
a2
1
= 0 ,
0
v01,B
ey2 ,B
0
= 1 ,
0
v cos
= v sin + a1
0
(7.72)
v02,B
v cos
= v sin a2 ,
0
(7.73)
are the distances from the center of gravity to the front and rear axle,
Fy = cS sy ,
where
cS
(7.74)
is dened by
Eq. (3.88). Because the vehicle is neither accelerated nor decelerated, the rolling condition
is fullled at each wheel
rD = eTx v0P .
Here,
rD
v0P
(7.75)
ex
vy = eTy v0P
(7.76)
and the rolling condition Eq. (7.75), the lateral slip can be calculated from
eTy v0P
,
sy = T
| ex v0P |
with
ey
(7.77)
labeling the unit vector in the lateral direction direction of the tire. So, the lateral
(7.78)
sy1 =
126
.
| cos (v cos ) + sin (v sin + a1 )
|
(7.79)
sy2 =
The yaw velocity of the vehicle
v sin a2
.
| v cos |
(7.80)
are
considered to be small
| a1 | |v| ;
| | 1
| a2 | |v|
(7.81)
|| 1 .
(7.82)
and
Because the side slip angle always labels the smaller angle between the velocity vector
and the vehicle longitudinal axis, instead of
v sin v
the approximation
v sin |v|
(7.83)
sy1 =
a1
v
+
|v|
|v|
(7.84)
a2
,
|v|
(7.85)
and
sy2 = +
where the consequences of Eqs. (7.81), (7.82), and (7.83) were already taken into consideration.
1, Eq. (7.71)
v
vC,B = |v| .
0
can be approximated by
(7.86)
0F,B
0
= 0 .
If the vehicle accelerations are also expressed in the vehicle xed frame
(7.87)
xF , yF , zF ,
one
0
= 0F,B vC,B + v C,B = v + |v| .
(7.88)
0
aC,B
127
Vehicle Dynamics
0F,B
0
= 0 ,
(7.89)
(7.90)
was used. The linear momentum in the lateral direction of the vehicle reads as
= Fy1 + Fy2 ,
m (v + |v| )
where, due to the small steering angle, the term
and
Fy1 cos
(7.91)
describes the vehicle mass. With Eq. (7.90) the angular momentum yields
= a1 Fy1 a2 Fy2 ,
where
Fy1 ,
(7.92)
names the inertia of vehicle around the vertical axis. With the linear description
of the lateral forces Eq. (7.78) and the lateral slips Eqs. (7.84), (7.85), one gets from
Eqs. (7.91) and (7.92) two coupled, but linear rst order dierential equations
cS1
m |v|
a1 cS1
a2
cS2
v
|v|
m |v|
|v|
a1
v
a2
a2 cS2
+
,
|v|
|v|
|v|
a1
v
+
|v|
|v|
(7.93)
(7.94)
cS1 + cS2
m |v|
a2 cS2 a1 cS1
| {z }
x
|
v
a2 cS2 a1 cS1
v cS1
m |v||v|
|v|
|v| m |v|
+
| {z } v a1 cS1
a21 cS1 + a22 cS2
x
|v|
|v|
{z
}
|
{z
A
B
.
|{z}
u
(7.95)
7.3.6 Stability
7.3.6.1 Eigenvalues
The homogeneous state equation
x = A x
128
(7.96)
xh (t) = x0 et
(7.97)
( E A) x0 = 0 .
One gets non-trivial solutions
x0 6= 0
(7.98)
for
det | E A| = 0 .
The eigenvalues
(7.99)
cS1 + cS2
m |v|
=
Av0
v
a2 cS2 a1 cS1
m |v||v|
|v|
2
2
a1 cS1 + a2 cS2
|v|
(7.100)
v 0.
The matrix in
(7.101)
1v0 =
cS1 + cS2
m |v|
and
2v0 =
The eigenvalues are real and always negative independent from the driving direction.
Thus, vehicles possess an asymptotically stable driving behavior at low speed!
v ,
Av
=
a2 cS2 a1 cS1
|v|
.
(7.102)
Using Eq. (7.102) one receives from Eq. (7.99) the relation
2v +
v a2 cS2 a1 cS1
= 0
|v|
(7.103)
129
Vehicle Dynamics
1,2v
When driving forward with
v > 0,
r
v a2 cS2 a1 cS1
.
=
|v|
(7.104)
(7.105)
holds. Then however, one eigenvalue is positive, and the system is unstable. Two zeroeigenvalues
1 = 0
and
2 = 0
a1 cS1 = a2 cS2 .
(7.106)
The driving behavior is indierent then. Slight parameter variations, however, can lead
to an unstable behavior. With
v > 0
or
(7.107)
the root argument in Eq. (7.104) becomes negative. Then, the eigenvalues
are imaginary, and disturbances lead to undamped vibrations. To avoid instability, highspeed vehicles have to satisfy the condition Eq. (7.107). The root argument in Eq. (7.104)
changes at backward driving its sign. Hence, a vehicle showing stable driving behavior at
forward driving becomes unstable at fast backward driving!
xst = const.
or
equations
A xst = B u .
With the elements from the state matrix
st
(7.108)
B,
= 0
(7.109)
(7.110)
m |v| |v|
|v|.
st =
130
v cS1 0
v a1 cS1 0
(7.111)
and
v cS1 0
st =
(7.112)
detD 6= 0,
(7.113)
a1
v
cS2 (a1 + a2 )
=
0 ,
|v| a + a + m v |v| a2 cS2 a1 cS1
1
2
cS1 cS2 (a1 + a2 )
a2 m v |v|
st
st =
a1 + a2 + m v |v|
At forward driving vehicles
a2 cS2 a1 cS1 0 .
cS1 cS2 (a1 + a2 )
(7.114)
(7.115)
v > 0, the steady state side slip angle starts with the kinematic
value
v0
st
=
v
a2
0
|v| a1 + a2
v0
st
=
and
v
0
a1 + a2
(7.116)
s
vst=0 =
a2 cS2 (a1 + a2 )
a1 m
(7.117)
the side slip angle changes the sign. Using the kinematic value of the yaw velocity
Eq. (7.115) can be written as
st =
v
a1 + a2
1
1 +
where
s
vch =
|v|
v
0 ,
(7.118)
v 2
vch
(7.119)
131
Vehicle Dynamics
radius of curvrature
200
0
150
r [m]
[deg]
-2
-4
-6
-8
-10
100
10
20
v [m/s]
m=700 kg;
=1000 kg m2 ;
30
a1 =1.2 m;
a2 =1.3 m;
a1*cS1/a2*cS2 = 0.66667
a1*cS1/a2*cS2 = 1
a1*cS1/a2*cS2 = 1.3333
50
0
40
10
cS1 = 80 000 N m;
20
v [m/s]
cS2
30
40
110 770 N m
= 73 846 N m
55 385 N m
R0 = 100 m
at
st =
v 0.
v
.
R
(7.120)
Some concepts for an additional steering of the rear axle were trying to keep the side
slip angle of the vehicle, measured at the center of the vehicle to zero by an appropriate
steering or controlling. Due to numerous problems, production stage could not yet be
reached.
0 =
a1 + a2
v 2 v a2 cS2 a1 cS1
+ m
.
R
R |v| cS1 cS2 (a1 + a2 )
(7.121)
The rst term is the Ackermann steering angle, which follows from Eq. (7.2) with the
wheel base
a = a1 + a2
tan 0 0 .
The Ackermann-steering angle provides a good approximation for slowly moving vehicles,
v 0. Depending on
v > 0, backward: v < 0),
a2 cS2 a1 cS1
the necessary steering angle diers from the Ackermann-steering angle at higher speeds.
The dierence is proportional to the lateral acceleration
ay =
132
v2
.
R
(7.122)
At
v > 0 the
a1 , a2
gravity
cS1 , cS2 .
understeering
0 > 0A
or
or
a1 cS1
<1
a2 cS2
neutral
0 = 0A
or
a1 cS1 = a2 cS2
or
a1 cS1
=1
a2 cS2
oversteering
0 < 0A
or
or
a1 cS1
>1
a2 cS2
st = 0, st = 0
Eq. (7.120), the equations of motion Eq. (7.91) and Eq. (7.92) can be dissolved for the
lateral forces
Fy1st =
Fy2st =
a2
v2
,
m
a1 + a2
R
or
a1
v2
m
a1 + a2
R
a1
Fy2st
.
=
a2
Fy1st
(7.123)
st
Fy1
= cS1 sst
y1
where
sst
yA1
and
sst
yA2
and
st
Fy2
= cS2 sst
y2 ,
(7.124)
label the steady state lateral slips at the axles. Now, from Eqs. (7.123)
st
Fy2
cS2 sst
a1
y2
= st =
a2
Fy1
cS1 sst
y1
or
sst
a1 cS1
y2
= st .
a2 cS2
sy1
< a2 cS2 )
(7.125)
during steady
state cornering the slip angles at the front axle are larger than the slip angles at the rear
st
st
axle, sy1 > sy2 . So, the steering tendency can also be determined from the slip angle at
the axles.
clin
S1 = cS Fz1
and
clin
S2 = cS Fz2 .
(7.126)
133
Vehicle Dynamics
G = mg
Fz1 =
a2
G
a1 + a2
.Fz2 =
and
a1
G
a1 + a2
(7.127)
a1 clin
S1 = a1 cS
a2
G
a1 + a2
(7.128)
a2 clin
S2 = a2 cS
a1
G.
a1 + a2
(7.129)
and
Thus, a vehicle with identical tires would be steering neutrally at a linear inuence of the
wheel load on the cornering stiness, because of
lin
a1 clin
S1 = a2 cS2
(7.130)
The lateral force is applied behind the center of the contact patch at the caster oset
v
distance. Hence, the lever arms of the lateral forces change to a1 a1
n and
|v| L1
v
a2 a2 + |v| nL1 , which will stabilize the vehicle, independently from the driving direction.
6
5
Fz [N ]
Fy [kN]
4
3
2
1
0
Fz [kN]
Figure 7.15: Lateral force
Fy
Fz
Fy [N ]
1000
758
2000
1438
3000
2043
4000
2576
5000
3039
6000
3434
7000
3762
8000
4025
At a real tire, a degressive inuence of the wheel load on the tire forces is observed,
Fig. 7.15. According to Eq. (7.92) the rotation of the vehicle is stable, if the torque from
the lateral forces
Fy1
and
Fy2
is aligning, i.e.
(7.131)
134
The
a1 < a 2
135
speed
Typical
80
60
40
20
0
-2
-4
5
3
2
1
0
4
3
2
1
0.2
0.4
0.6
0.8
lateral acceleration [g]
0.2
0.4
0.6
0.8
lateral acceleration [g]
R = 100 m
The vehicle is under-steering and thus stable according to Eq. (7.121) with Eq. (7.122).
The inclination in the diagram steering angle versus lateral velocity decides about the
steering tendency and stability behavior.
136
The nonlinear inuence of the wheel load on the tire performance is here used to design
a vehicle that is weakly stable, but sensitive to steer input in the lower range of lateral
acceleration, and is very stable but less sensitive to steer input in limit conditions.
With the increase of the lateral acceleration the roll angle becomes larger. The overturning
torque is intercepted by according wheel load dierences between the outer and inner
wheels. With a suciently rigid frame the use of an anti roll bar at the front axle allows
to increase the wheel load dierence there and to decrease it at the rear axle accordingly.
Thus, the digressive inuence of the wheel load on the tire properties, cornering stiness
and maximum possible lateral force, is stressed more strongly at the front axle, and the
vehicle becomes more under-steering and stable at increasing lateral acceleration, until it
drifts out of the curve over the front axle in the limit situation.
Problems occur at front driven vehicles, because due to the demand for traction, the front
axle cannot be relieved at will.
Having a suciently large test site, the steady state cornering maneuver can also be
carried out at constant speed. There, the steering wheel is slowly turned until the vehicle
reaches the limit range. That way also weakly motorized vehicles can be tested at high
lateral accelerations.
40
30
20
10
0
0.2
0.4
0.6
time [s]
0.8
137
12
0.5
10
0.4
0.6
0.3
0.2
0.1
6
4
2
2.5
0.5
side slip angle [deg]
Vehicle Dynamics
2
1.5
1
0.5
0
0
-0.5
-1
-1.5
-2
[t]
v = 100 km/h
The vehicle under consideration behaves dynamically very well, Fig. 8.3. Almost no overshoots occur in the time history of the roll angle and the lateral acceleration. However,
small overshoots can be noticed at yaw the velocity and the sideslip angle.
z = z(x, y)
z1 = z1 (s1 ) ,
and
z2 = z2 (s2 )
(8.1)
z = z(x, y)
138
z = z(s) ;
(8.2)
0.05
0.04
0.03
0.02
0.01
0
-0.01
-0.02
-0.03
-0.04
-0.05
0
10
20
30
40
50
60
70
80
90
100
z = z(s)
is completely
1
mz = lim
s 2s
Zs
z(s) ds
(8.3)
s
and the correlation function
Zs
1
Rzz () = lim
s 2s
z(s) z(s ) ds .
(8.4)
s
A vanishing mean value
mz = 0
Rzz () = Rzz () ,
and
1
Rzz (0) = lim
s 2s
Zs
z(s)
(8.5)
2
ds
(8.6)
s
describes the variance of
zs .
Stochastic track irregularities are mostly described by power spectral densities (abbreviated by psd). Correlating function and the one-sided power spectral density are linked by
the Fourier-transformation
Z
Rzz () =
Szz () cos() d
(8.7)
0
where
denotes the space circular frequency. With Eq. (8.7) follows from Eq. (8.6)
Z
Rzz (0) =
Szz () d .
(8.8)
139
Vehicle Dynamics
Thus, the psd gives information, how the variance is compiled from the single frequency
shares.
The power spectral densities of real tracks can be approximated by the relation
Szz () = S0
w
,
(8.9)
notable irregularities in the short or long wave spectrum. At real tracks, the reference6
psd S0 lies within the range from 1 10
m3 to 100 106 m3 and the waviness can be
approximated by
w = 2.
highway: S 0=1*10
-5
m ; w=2
1000
1000
500
500
-2
[deg] 2
-2
m ; w=2
[deg] 2
v = 90 km/h
are
displayed in Fig. 8.5. The track quality is reected in the amount of steering actions. The
steering activity is often used to judge a vehicle in practice.
140
vehicle
mass
[kg]
center of gravity
[m]
empty
12 500
fully laden
18 000
inertias
[kg m2 ]
12 500
0
0
0
155 000
0
0
0
155 000
15 400
0
250
0
200 550
0
250
0
202 160
has a double wishbone single wheel suspension. The twin-tire rear axle with the track
i
o
widths sh = 2.152 m and sh = 1.492 m is guided by two longitudinal links and an a-arm.
The air-springs are tted to load variations via a niveau-control.
0
steer angle [deg]
100 m-Radius.
pied vehicle is slightly more understeering than the empty one. The higher wheel loads
cause greater tire aligning torques and increase the degressive wheel load inuence on the
increase of the lateral forces. Additionally roll steering at the rear axle occurs.
141
Vehicle Dynamics
steer angle
250
LW
[deg]
vehicle course
200
200
[m]
150
150
100
100
50
50
100
50
-100
0
[m]
100
100
50
fully occupied
Both vehicles can not be kept on the given radius in the limit range. Due to the high position of the center of gravity the maximal lateral acceleration is limited by the overturning
hazard. At the empty vehicle, the inner front wheel lift o at a lateral acceleration of
ay 0.4 g
ay 0.35 g .
v = 80 km/h
can be seen in
Fig. 8.8. To achieve comparable acceleration values in steady state condition, the step
steer input was done at the empty vehicle with = 90 and at the fully occupied one
with = 135 . The steady state roll angle is 50% larger at the fully occupied bus than
at the empty one. By the niveau-control, the air spring stiness increases with the load.
Because the damper eect remains unchanged, the fully laden vehicle is not damped as
well as the empty one. This results in larger overshoots in the time histories of the lateral
acceleration, the yaw angular velocity, and the sideslip angle.
142
yaw velocity
10
0.2
0.1
2
0
roll angle
[deg]
[deg]
0
-1
2
0
Z [deg/s]
0.3
-2
0
[s] 6
- - coach empty,
[s] 6
100 m
radius are
143
Vehicle Dynamics
10
steer angle
5
0
-5
-10
-5
LW
100
0
lateral motion [cm]
[deg]
roll angle
trailing arm
[Grad]
4
3
50
2
1
0.2
0.4
0.6
0.8
2
0
0.2
0.4
0.6
0.8
lateral acceleration a y [g]
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8
lateral acceleration a y [g]
trailing
arm
The vehicle with a trailing arm rear axle is, compared to the serial car, more understeering.
The lack of roll support at the rear axle also causes a larger roll angle.
144
Index
Damping rate, 76
Deviation, 13
Aerodynamic forces, 96
Disturbance-reaction problem, 84
Air resistance, 96
Down forces, 96
Downhill capacity, 97
Anti-lock-system, 102
Driver, 3
Auto-correlation, 13
Driving safety, 72
Double wishbone, 7
McPherson, 7
Multi-link, 7
Axle load, 95
Axle suspension
Solid axle, 55
Twist beam, 56
Bend angle, 113, 116
Brake pitch angle, 103
Brake pitch pole, 108
Braking force distribution, 100
Camber angle, 6, 24
Camber compensation, 121, 124
Camber slip, 49
Caster, 8, 9
Characteristic speed, 131
Eective value, 13
Eigenvalues, 128
Environment, 4
First harmonic oscillation, 63
Fourier-approximation, 64
Frequency domain, 63
Friction, 97
Front wheel drive, 98, 118
Generalized uid mass, 70
Grade, 95
Hydro-mount, 69
Climbing capacity, 97
Kingpin, 7
Comfort, 72
Kingpin Angle, 8
Contact point, 24
Cornering resistance, 117, 118
Vehicle Dynamics
Ljapunov equation, 84
State vector, 88
Load, 4
Oversteering, 133
Lever arm, 57
Rack, 57
Bore slip, 52
Camber angle, 24
Reference frames
Camber inuence, 48
Ground xed, 5
Characteristics, 52
Inertial, 5
Circumferential direction, 24
Vehicle xed, 5
Composites, 19
Contact forces, 20
Ride comfort, 83
Contact patch, 20
Ride safety, 83
Contact point, 23
Road, 10, 23
Contact torques, 20
Cornering stiness, 41
Deection, 26
Deformation velocity, 31
Development, 19
Dynamic oset, 41
Dynamic radius, 32, 33
Safety, 72
Friction coecient, 45
Lateral direction, 24
Lateral force, 20
Spring rate, 78
Stability, 128
Lateral slip, 40
Lateral velocity, 31
State matrix, 88
ii
Lift o, 88
125
Vertical dynamics, 72
Waviness, 140
Wheel
Angular velocity, 50
Wheel base, 109
Wheel load, 20
Wheel loads, 94
Wheel rotation axis, 5
Wheel Suspension
Semi-trailing arm, 143
Single wishbone, 143
Trailing arm, 143
Wheel suspension
Central control arm, 56
Double wishbone, 55
McPherson, 55
Multi-Link, 55
Semi-trailing arm, 56
SLA, 56
Yaw angle, 112, 115
Yaw velocity, 126
Tire Model
Kinematic, 109
Linear, 133
TMeasy, 52
Toe angle, 6
Toe-in angle, 6
Track, 23
Track curvature, 114
Track normal, 5
Track radius, 114
Track width, 109, 121
Trailer, 112, 115
Understeering, 133
Variance, 13
Vehicle, 3
Vehicle comfort, 72
Vehicle dynamics, 2
iii