III European Conference on Computational Mechanics

Solids, Structures and Coupled Problems in Engineering

C.A. Mota Soares et.al. (eds.)
Lisbon, Portugal, 58 June 2006
FIRST ORDER TIRE DYNAMICS
Georg Rill
University of Applied Sciences Regensburg
Galgenbergstr. 30
D-93053 Regensburg
e-mail: GeorgRill@aol.com
Keywords: Tire Modeling, Tire Dynamics, Stick Slip Effects
Abstract. In this paper a method is presented where the rst order tire dynamics are generated
by a Taylor-Expansion of the steady state forces and torques. Thus, relaxation lengths which
include the wheel load and slip dependencies are automatically generated from the steady state
tire properties. Slight model modications make it possible to simulate even stick slip effects
during parking maneuvers. The results of this simple but effective approach correspond quite
well with measurements.
1
G. Rill First Order Tire Dynamics
1 Introduction
1.1 Tire Modeling
For the dynamic simulation of on-road vehicles, the model-element tire/road is of special
importance, according to its inuence on the achievable results. It can be said that the suf-
cient description of the interactions between tire and road is one of the most important tasks
of vehicle modeling, because all the other components of the chassis inuence the vehicle dy-
namic properties via the tire contact forces and torques. Therefore, in the interest of balanced
modeling, the precision of the complete vehicle model should stand in reasonable relation to
the performance of the applied tire model. At present, two groups of models can be identied,
handling models and structural or high frequency models, [4].
Structural tire models like RMOD-K [6] or FTire [1] are very complex. These models are
computer time consuming and they need a lot a data. Usually, they are used for stochastic
vehicle vibrations occurring during rough road rides and causing strength-relevant component
loads, [8].
Comparatively lean tire models are suitable for vehicle dynamics simulations, while, with
the exception of some elastic partial structures such as twist-beam axles in cars or the vehicle
frame in trucks, the elements of the vehicle structure can be seen as rigid.
Handling models like the MF-formula [7] or TMeasy [2] rely also on measured and observed
force-slip characteristics. This class of tire models is characterized by an useful compromise
between user-friendliness, model-complexity and efciency in computation time on the one
hand, and precision in representation on the other hand.
1.2 Steady State Tire Forces and Torques
In any point of contact between the tire and the road surface normal and friction forces are
transmitted, Fig. 1. For the calculation of the contact patch geometry the tire is considered as
F
x
longitudinal force
F
y
lateral force
F
z
vertical force or wheel load
T
x
tipping torque
T
y
rolling resistance torque
T
z
self aligning and bore torque
F
x
F
y
F
z
T
x
T
y
T
z
e
yR
Figure 1: Tire forces and torques
a rigid body. Then, the tire deformation, the orientation of the local contact area, the location
of the contact point, and the contact point velocities can be calculated from the momentary
state of the wheel rim and the description of the road surface. Adding the knowledge or at
least an approximation about the tire elasticity results in the wheel load F
z
and the tipping
torque T
x
. The rolling resistance torque T
y
which is less important for vehicle dynamics may
be approximated by a rolling resistance lever, [5].
Within handling models the steady state tire forces in longitudinal and lateral direction are
2
G. Rill First Order Tire Dynamics
approximated by appropriate functions
F
S
x
= F
S
x
(s
x
, s
y
, . . .) (1)
F
S
y
= F
S
y
(s
x
, s
y
, . . .) (2)
which mainly depend on the longitudinal and lateral slip s
x
and s
y
. The steady state torque
T
S
z
around an axis perpendicular to the local road plane consists of the self aligning torque
T
S
S
= c
o
F
S
y
and the bore torque T
S
B
T
S
z
= c
o
F
S
y
+ T
S
B
(s
B
. . .) , (3)
where c
o
= c
o
(s
y
) names the tire caster and s
B
denotes a bore slip.
1.3 Simple Dynamic Extension
Measurements show that the dynamic reaction of the tire forces and torques to disturbances
can be approximated quite well by rst order systems. Then, the dynamic tire forces F
D
x
, F
D
y
and the dynamic tire torque T
D
z
are given by rst order differential equations

x

F
D
x
+ F
D
x
= F
S
x
(4)

y

F
D
y
+ F
D
y
= F
S
y
(5)

T
D
z
+ T
D
z
= T
S
z
(6)
which are driven by the the steady values F
S
x
, F
S
y
and T
S
z
. The tread particles of a rolling tire
move with the transport velocity r
D
|| through the contact patch, where r
D
and denote the
dynamic rolling radius and the angular velocity of the wheel. Now, time constants
i
, can be
derived from so called relaxation lengths r
i

i
=
r
i
r
D
||
i = x, y, . (7)
But it turned out that these relaxation lengths are functions of the longitudinal and lateral slip
s
x
, s
y
and the wheel load F
z
, Fig. 2. Therefore, constant relaxation lengths will approximate
the real tire behavior in zero order approximation only. An appropriate model for the dynamic
tire performance would be of great advantage because then, the cumbersome task of deriving
the relaxation lengths from measurements can be avoided.
2 Force Dynamics
2.1 Lateral Force
The lateral force F
y
acting in the contact patch deects the tyre in lateral direction, Fig. 3.
The rst order approximation
F
y
(v
y
+ y
e
)
. .
F
D
y
F
y
(v
y
)
. .
F
S
y
+
F
y
v
y
y
e
(8)
takes the lateral tyre deection y
e
into account. Within TMeasy [2] the lateral slip is dened by
s
y
=
v
y
r
D
||
, (9)
3
G. Rill First Order Tire Dynamics
1 2 3 4 5 6 8 9 7 10 0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
F
z
= 6 kN
F
z
= 4 kN
F
z
= 2 kN
slip angle [
o
]
r
y
[m]
Figure 2: Measured lateral force relaxation length for a typical passenger car tire, [3]
where v
y
describes the sliding velocity of the contact point in lateral direction, r
D
|| denotes
the velocity with which the tread particles are transported through the contact area. Now, the
derivative of the lateral force F
y
with respect to the lateral component v
y
of the contact point
velocity can be calculated from
F
y
v
y
=
F
y
s
y
s
y
v
y
=
F
y
s
y
1
r
D
||
, (10)
where F
y
/s
y
can easily be derived from the lateral force characteristics F
y
= F
y
(s
y
, . . .).
If c
y
and d
y
denote the lateral stiffness and the lateral damping property of the tire then, the
d
y
c
y
F
y
v
y
Q P
y
e
tire
rim
Figure 3: Lateral Tyre Deection
dynamic tire force F
D
y
= F
y
(v
y
+ y
e
) will also result from
F
D
y
= c
y
y
e
+ d
y
y
e
. (11)
Inserting (10) into (8) and combining it with (11) yields a rst order differential equation for
the lateral tire deection y
e
F
S
y
+
F
y
s
y
1
r
D
||
y
e
= c
y
y
e
+ d
y
y
e
(12)
4
G. Rill First Order Tire Dynamics
which can be written as
1
c
y
_
d
y
+
F
y
s
y
1
r
D
||
_
. .

y
y
e
+ y
e
=
F
S
y
c
y
. (13)
According to (7) the relaxation length for the lateral tire deection and hence, for the lateral tire
force is now given by
r
y
= r
D
||
y
= r
D
||
1
c
y
_
d
y
+
F
y
s
y
1
r
D
||
_
= r
D
||
d
y
c
y
+
1
c
y
r
D
||
r
D
||
F
y
s
y
. (14)
This simple model approach already leads to a relaxation length which is automatically adapted
to the tire parameter. The relaxation length r
y
for the lateral tire deection depends on the wheel
load F
z
and it is a function of the lateral slip s
y
or the slip angle = arctan s
y
, Fig. 4. This
8
6
4
2
0
[kN]
slip angle [
o
]
0 5 10 15
slip angle [
o
]
0 2 4 6 8 10
[m]
0.8
0.6
0.4
0.2
0
F
z
=2 kN
F
z
=4 kN
F
z
=6 kN
F
z
=8 kN
Relaxation length r
y
Lateral force characteristics F
y
Figure 4: Lateral force characteristics and computed relaxation length for a typical passenger car tire (S205/50
R15, 6J) with c
y
= 160 000 N/m, d
y
= 600 N/(m/s), r
D
|| = 60 km/h
approach needs steady state tire characteristics only. By introducing an appropriate weighting
function this simple approach can be adapted to measured relaxation lengths.
2.2 Longitudinal Force
Similar to the lateral force dynamics the rst order approximation
F
x
(v
x
+ x
e
)
. .
F
D
x
F
x
(v
x
)
. .
F
S
x
+
F
x
v
x
x
e
(15)
takes the longitudinal tyre deection x
e
into account. The derivative of the longitudinal force
F
x
with respect to the longitudinal component v
x
of the contact point velocity can be calculated
from
F
x
v
x
=
F
x
s
x
s
x
v
x
=
F
x
s
x
1
r
D
||
(16)
5
G. Rill First Order Tire Dynamics
where the longitudinal slip denition
s
x
=
r
D
v
x
r
D
||
(17)
was used to generate the derivative of the longitudinal slip with respect to the longitudinal
component of the contact point velocity. The partial derivative of the longitudinal force F
x
with
respect to the longitudinal slip s
x
will be provided by the steady state tire characteristics.
If c
x
and d
x
denote the longitudinal stiffness and the longitudinal damping of the tyre, the
dynamic tire force F
D
x
= F
x
(v
x
+ x
e
) can also be calculated by
F
D
x
= c
x
x
e
+ d
x
x
e
. (18)
Combining (15) and (18) results in a rst order differential equation for the longitudinal tyre
deection x
e
_
d
x
+
F
x
s
x
1
r
D
||
_
x
e
= F
S
x
c
x
x
e
. (19)
Again, the dynamics of the longitudinal tire deection and hence, of the longitudinal tire force
can be characterized in the space domain by the relaxation length
r
x
= r
D
||
d
x
c
x
+
1
c
x
F
x
s
x
. (20)
Via the partial derivative F
x
/s
x
the relaxation length depends on the wheel load F
z
and on
the longitudinal slip s
x
, Fig. 5.
[m]
0.8
0.6
0.4
0.2
0
1.0
F
z
=2 kN
F
z
=4 kN
F
z
=6 kN
F
z
=8 kN
slip [%]
0 2 4 6 8 10
8
6
4
2
0
[kN]
10
slip [%]
0 20 40 60 80 100
Relaxation length r
x
Longitudinal force characteristics F
x
Figure 5: Longitudinal force characteristics and computed relaxation length for a typical passenger car tire
(S205/50 R15, 6J) with c
y
= 220 000 N/m, d
y
= 700 N/(m/s), r
D
|| = 60 km/h
3 Torque Dynamics
3.1 Self Aligning Torque
The self aligning torque is generated by the lateral force. By neglecting a possible dynamics
of the tire offset the dynamic self aligning torque can be approximated by
T
D
S
= c
o
F
D
y
, (21)
where c
o
denotes the steady state tire offset and the dynamics of the lateral tire force F
D
y
char-
acterized by the relaxation length (14) also denes the dynamics of the self aligning torque.
6
G. Rill First Order Tire Dynamics
3.2 Bore Torque
Similar to the longitudinal and lateral slip a bore slip can be dened by
s
B
=
r
B

n
r
D
||
, (22)
where r
B
denotes the bore radius of the contact patch and
n
describes the component of the
wheel angular velocity around an axis normal to the road. Such bore motions will occur if the
tire is cambered or if the wheel is steered. Now, the steady state bore torque can roughly be
modeled by
T
S
B
= r
B
dF
0
s
B
, with

T
S
B

r
B
F
max
, (23)
where dF
0
and F
max
are the initial inclination and the maximum value of the generalized tire
force characteristics which is used in TMeasy [2] to generate combined tire forces. Similar to
the tire forces the dynamic bore torque can be modeled in a rst approximation by
T
D
B
= T
B
(
n
+
t
) = T
S
B
+
T
S
B

t
, (24)
where
t
describes the torsional tire deection. Using (23) and (22) one gets
T
D
B
= T
S
B
+ r
B
dF
0
r
B
r
D
||

t
. (25)
But, the dynamic bore torque can also be calculated via the torsional tire stiffness c
t
and the
torsional tyre damping d
t
T
D
B
= c
t

t
+ d
t

t
. (26)
These two equations can be combined to a rst order differential equation for the torsional tire
deection
t
_
d
t
+
r
2
B
dF
0
r
D
||
_

t
= T
S
B
c
t

t
(27)
Hence, the relaxation length
r

= r
D
||
d
t
c
t
+
1
c
t
r
2
B
dF
0
(28)
characterizes the dynamics of the torsional tire deection and the bore torque. In this simple
approach r

depends only on the wheel load F

z
but, this corresponds quite well with measure-
ments, Fig.6
3.3 Parking Torque
Equation (27) is not yet ready to describe steering motions at stand still where 0 will
hold. Multiplying (27) with r
D
|| yields
_
r
D
|| d
t
+ r
2
B
dF
0
_

t
= r
D
|| T
S
B
r
D
|| c
t

t
. (29)
Using the rst part of (23) and (22) results in
_
r
D
|| d
t
+ r
2
B
dF
0
_

t
= r
2
B
dF
0

n
r
D
|| c
t

t
. (30)
7
G. Rill First Order Tire Dynamics
r

[
m
m
]
800
600
400
200
0
0 5 10 15 20
computed
Slip Angle [
o
]
F
z
= 1 kN
F
z
= 3 kN
F
z
= 5 kN
F
z
= 7 kN
r

[
m
m
]
800
600
400
200
0
0 5 10 15 20
measured
Slip Angle [
o
]
F
z
= 3 kN
F
z
= 5 kN
Figure 6: Measured [3] and computed bore torque relaxation length (c
t
= 9 000 Nm/rad)
Now, at stand still ( = 0) the simple differential equation

t
=
n
r
D
|| c
t

t
(31)
remains which means that the torsional tire deection
t
is increased or decreased as long as
steering motions
n
= 0 are performed. But the differential equation (30) does not yet take
into account that the steady state bore torque is limited to a maximum value. Applying this
restriction to the steady state part of the dynamic bore torque results in

T
S
B

T
max
B
or |c
t

t
| T
max
B
, (32)
where T
max
B
= r
B
F
max
names the maximum bore torque. Combining (32) with (30) nally
yields

t
=
r
2
B
dF
0

n
+ r
D
|| c
t

t
r
D
|| d
t
+ r
2
B
dF
0
;

t
=
_

t
if |c
t

t
| T
max
B
,
0 if |c
t

t
| > T
max
B
&
t

t
> 0 .
(33)
Fig. 7 shows the results of a low frequency sine steer input at stand still. Of course this
simple rst order bore torque dynamics results only in a rough approximation to the real tire
behavior. But this approach is based on steady state tire properties and generates and automatic
transition from stand still (v
t
= 0) to normal driving situations (v
t
= 0).
4 Conclusion
In this paper a method is presented where the rst order tire dynamics is generated by a
Taylor-Expansion of the steady state forces and torques. Thus, relaxations lengths which in-
clude the wheel load and slip dependencies are automatically generated from the steady state
tire properties. Slight model modications make it possible to simulate stick slip effects during
parking maneuvers. The results of this simple but effective approach correspond quite well with
measurements.
8
G. Rill First Order Tire Dynamics
-20 -10 0 10 20
-400
-200
0
200
400
T
B

[
N
m
]

wheel
[
o
]
measured
computed
Figure 7: Measured [3] and computed bore torque at stand still
REFERENCES
[1] Flexible Ring Tire Model Documentation and Users Guide. Cosin Consulting 2004,
http://www.ftire.com.
[2] Hirschberg, W; Rill, G. Weinfurter, H.: User-Appropriate Tyre-Modeling for Vehicle Dy-
namics in Standard and Limit Situations. Vehicle System Dynamics 2002, Vol. 38, No. 2,
pp. 103-125. Lisse: Swets & Zeitlinger.
[3] van der Jagt, P.: The Road to Virtual Vehicle Prototyping; new CAE-models for acceler-
ated vehicle dynamics development. PhD-Thesis, Tech. Univ. Eindhoven, Eindhoven 2000,
ISBN 90-386-2552-9 NUGI 834.
[4] Lugner, P.; Pacejka, H.; Pl ochl,M.: Recent advances in tyre models and testing procedures.
Vehicle System Dynamics, Vol. 43, No. 67, JuneJuly 2005, 413436.
[5] Mitschke, M.; Wallentowitz, H.: Dynamik der Kraftfahrzeuge. 4. Auage. Springer-Verlag
Berlin Heidelberg 2004.
[6] Oertel, Ch.; Fandre, A.: Ride Comfort Simulations an Steps Towards Life Time Calcula-
tions; RMOD-K and ADAMS. International ADAMS User Conference, Berlin 1999.
[7] Pacejka, H.B.: Tyre and Vehicle Dynamics. Oxford: Butterworth-Heinemann, 2002.
[8] Riepl, A.; Reinalter, W.; Fruhmann, G.: Rough Road Simulation with tire model RMOD-K
and FTire. In: Proc. of the 18th IAVSD Symposium on the Dynamics of vehicles on Roads
and on Tracks. Kanagawa, Japan, 2003. Taylor & Francis, London UK.
9