Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Model Tire Vibration

Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

Masahiro Takayama I and Koichi Yamagishi 1

Simulation Model of Tire Vibration 2

REFERENCE: Takayama, M. and Yamagishi, K., "Simulation Model of Tire Vibration",


Tire Science and Technology, Vol. 11, Nos. 1-4, January - December, 1984, pp. 38-49.

ABSTRACT: A mass-spring tire model having five degrees of freedom is used to analyze
the tangential and radial axial forces that result from a tire hitting a cleat. The belt and tread
region is modeled by a rigid ring. Deflections from the cleat are absorbed by a line spring
and those from the smooth portion of the test wheel (the ground) are absorbed by a plane
spring; both are attached to the rigid ring. Caicutated results agreed well with experimental
results.

KEY WORDS: Tire vibrations, tire model, cleat envelopment, axial forces.

Nomenclature
C = Envelopment coefficient
G = D a m p i n g coefficient
i= 1 Rigid ring m o t i o n
i= 2 Line deformation
i-- 3 Tread compression
i-- 4 Plane deformation
i-- 5 & 6 Tread shear deformation
D :- Torsion damping coefficient
G = Torsion spring of sidewall
f i = Cleat reaction force
fi = Contact force from ground
FR = Radial axial force (vertical)
Fx = Tangential axial force (horizontal)
FRpv = Max. amplitude of FR
Frvp = Max. amplitude of Fx

IBridgestone Tire and Rubber Co., Ltd., Ogawa Higashi-Cho, Kodaira-Shi, Tokyo 187, Japan.
2Presented at the inaugural meeting of The Tire Society, March 25-26, 1982 at the University of
Akron, Akron, Ohio.
38
TAKAYAMA AND YAMAGISHI ON TIRE VIBRATION 39

h= Cleat height
L= Footprint contact length
M1 = Mass of ring
M2= Equivalent tread mass deformed by cleat
F = Radius of ring

t= Time
V= Tire speed
W= Load on tire
X1 = Tangential (horizontal) displacement of ring
X2= Radial (vertical) displacement of ring
X3= Circumferential displacement of M2
Torsion angle of ring
0= Angle of cleat rotation
00= Initial contact angle of cleat

Fundamental Assumptions

Tire forces and vibrations are transmitted through the suspension system of a
vehicle to the driver and passengers. In the present study, however, we assume for
simplicity a fixed axle (Fig. l) and neglect any effects on or by the suspension
system. Two-dimensional in-plane forces and displacements are assumed to
occur when a loaded tire roils on a surface. The surface is first considered to be
free of obstacles, then to include a cleat. Displacements and forces out of the
plane of the tire are neglected. The wheel is assumed to rotate at the same angular
velocity as that of the tire; there is no wind-up. Centrifugal forces are also
neglected.
Inputs from a Cleat and from a Smooth Surface
A dynamic tire model must include the envelopment of a cleat by the tire.
Majcher [1] did this well by using B6hm's model of an elastic ring on an elastic
foundation [6]. In the present investigation, however, an experimentally
determined cleat force is used as input force. The simple static tire contact
problem can be solved by starting with B6hm's model [2], then extending it to
solve the more general rolling vibration problem.
We have used the model in Fig. 1, where a displacement caused by the cleat is
defined as a line displacement and that caused by a smooth surface is a plane
displacement. Takebayashi [3] expressed a plane displacement as a function of a
line displacement, which led to much simpler expressions for combined
displacements. Tangential displacement of the tire by the cleat is assumed to be
negligibly small.
Various solutions for the eigenvalue problem of tires have been obtained from
models based on an elastic ring with an elastic foundation [4,5,6] as well as from
more sophisticated models [7,8].
40 TIRE SCIENCE TECHNOLOGY

(Cleat Height) h ~ . ~ / ~ . . -
L_J L5
(Cleat Force) fc fs (Contact Force)

FIG. l -- Cleat envelopment model.

Transmissibility of a Tire
The transmissibility of a tire may be defined as the ratio of its output force
(going to the axle) to its input force (coming from its ground contact). For its
analysis we extend the five degrees of freedom of a mass-spring system that was
proposed by Takebayashi [3] to match the tire engineering considerations shown
in Fig. 2.

FIG. 2 -- Tire simulation model having five degrees of freedom.

The region of the tread and belt is modeled as a rigid ring for which vibrational
modes higher than the second are neglected. Deflections caused by a cleat are
absorbed by line or plane springs attached to the rigid ring. The following
vibrations and motions are also considered:
1. Torsional vibration because of its importance to tangential axle force
T A K A Y A M A A N D Y A M A G I S H I O N TIRE V I B R A T I O N 41

2. Coupling among radial, tangential, and torsional movements


3. Tread shear spring at the cleat because of its relation to riding comfort
The system thus has seven springs as identified in the nomenclature section: ring
spring K1 for rigid movement, plane spring K4 for tire deflection on smooth
surfaces, line springs K2 and K3 for tire deflections on a cleat, tire torsion spring G,
and tread shear springs K5 and/(6, which have equal magnitudes. The categories
of springs are illustrated in Fig. 3, which also suggests measurements for
determination of the various spring constants. The tire load is represented by W
and the torsional force by T.

Ring Spring K1 Plone Spring K4

Line Spring Kz Torsion Spring G

FIG. 3 -- Determination o f spring constants.

Determination of Cleat Envelopment Index


A cleat envelopment index E n may be defined as
En : & / 6 o (1)
Approximations of & and 6o are illustrated in Fig. 4 where the dotted line
represents the shape of the undeflected tread. More precise definitions and
methods for determination of 61 and 60 will be developed later.

FIG. 4 -- Geometry of 6~ and 6o.


42 TIRE SCIENCE TECHNOLOGY

The envelopment index is a function of the position of the cleat along the tire
footprint. It will follow the shape shown in Fig. 5 if the spring constants for both
the line and plane springs are independent of deflection; it will increase linearly
from the point of contact with the cleat to a sharp maximum then decrease
linearly to the point of cleat exit. This dependence of En on the position of the
cleat along the X-axis is recognized by writing the index as En(X).

En(X) = Cx when 0 <~ x <_ L~ 2 (2)


En(X) = - C x + En(L/2) when L~ 2 <-- x <_ L

Methods for determining C by static experiments will be shown later. Such


experiments have confirmed the linearity of En as a function of X. Figure 6 shows
the relationship as the cleat proceeds from the point of entry to the footprint
center.

x.

I' L/a ' "' */a -I

FIG 5 -- Determination ofEn (X) as a function of footprint length L.

i/
E.. I/7
!yol
D
Leoding - Just Under
Edge X the Lood

FIG. 6 -- Experimental value of En as a function o f X.

When a tire traverses a cleat at very low speed the distribution of radial force
along the footprint has the two peaks shown in thef~ +f~ trace of Fig. 7. The force
on the cleat is represented by.l; and the ground contact force is represented bye.
The distribution of each of these two is shown in the bottom part of Fig. 7. The
trace for the cleat force is concave downward with a generally parabolic shape,
while that for the contact force is concave upward with essentially a two-
component linear shape.
TAKAYAMA AND YAMAGISHI ON TIRE VIBRATION 43

Cleot Force fc ContGct Force fs

FIG. 7 - - Envelopment characteristics o f tires. Ordinates are kilograms offorce. Multiply by 9.8
to convert to newtons, by 2.2 to convert to pounds force.

Differential equations 3 through 7 were obtained by applying basic dynamic


principles to the five-degree-of-freedom model in Fig. 2 and using the symbolism
given in the nomenclature section.

MIXI + (C1 + C2c0s20 + Cssin20) J(I + (C2sinOcosO - CscosOsinO) )(2


- Cssin0J(3 - C2cos0X4 + Csrsin0~
+ (K1 + K2cos20 + Kssin20) X1 + (K2sinOcosO - KscosOsinO) X2
- KssinOX3 - K 2 c o s O X . + K s r s i n O ( = 0 (3)
MIX2 + (C1 + C4 + C2sin20 + CscosZ0) J(2 + Cscos0J(3
- C2sin0J(4 - CsrcosO" "( + (C2sinOcosO - CssinOcosO) J(1
C46o + ( K I + K4 + K2sin20 + Kscos20) Xz + Kscos0 9 )(2
-

- K2sinOX4 - K s r c o s O " ( + (K2sinOcosO - KssinOcosO) X~


K46o : 0
- (4)
M2)(3 + ( G + C 6 ) J ( 3 - Csr~: - Cssin0JG + Gcos0J(2
+ C6c0s031 + (/<5 + K6) X3 - K s r " ( - K s s i n O X i
+ KscosOX2 + K6cos061 = 0 (5)
M2)(4 + (C2 + C3) )(4 -C2cos0JG - C2sin0J(2 - C3sind61
+ (K2 +/<3) X4 - K2cosOXl - K2sinOX2 - K3sinO& = 0 (6)
M l r i ( + ( D + C s r 2) "~ + CsrsinOJ(l - Csrcos0J(2 - CsrJ(3
+ ( G + K s r 2) ~ + KsrsinOX1 - KsrcosOX2 - K s r X 3 = 0 (7)

We must now determine values for & and 60.


Determination of&
An approximate value for 61, obtained geometrically from Fig. 8, is given by
~1 -- r sin 0 - (r - h - W / K4) (8)
44 TIRE SCIENCE TECHNOLOGY

where 0o --< 0 _< (rr - 0o)


h : height of the cleat
W = load on the tire

FIG. 8 -- Determination of&.

Determination o f 6o
Equations 2 are rewritten in Eqs 9 so as to express the envelopment index as a
function of 0 rather than of X
En (0) = C (0o O) / (rr/2 - 0o) when 0 % 0 _< rr/2
= C (0 n-+00)(n-/2 - 00) when rr/2 _< 0 --< (rr - 00) (9)
Then 60 = En (0) [ (x4 - xlcos0 - x2sin0) sin0 - (x3 - xtsin0 + x2cos0 - (r) cos0]
(lO)
The cleat angles are given by
O= V, / ( r - W/K4) (11)
and O0 : sin-~ ( r - h - W/K4)/r (12)
while the axial forces are given by
Fr : KIX2 + C1-,~2 (13)
and Ft : Klx~ + ClXl - G ( / r - D'~/r (14)
We can now set the time derivatives 2, Y, and ~ equal to zero in Eqs 3 through 7
and solve for the axial forces in the almost static case. Then C in Eqs 9 is
determined so as to agree with experiment. Since these differential equations are
non-linear, the R u n g e - K u t t a method is applied and axial forces are obtained
numerically.

Numerical Results

A 165SR13 tire was chosen to illustrate the numerical calculations. The


following properties and conditions were used:
M1 = 4 X 1 0 -4 kgfsZ/mm = 3.92 kg mass : 8.64 lb
M2 =5x10 -5 kgfs2/mm = 0.49 kg mass = 1.08 lb
K~ : 62.6 k g f / m m : 614 k N / m = 3506 1b/in.
K2 : 7.6 k g f / m m = 74.5 k N / m = 426 lb/in.
K3 : 55 k g f / m m = 539 k N / m : 3080 lb/in.
K4 = 16.3 k g f / m m : 160 k N / m : 913 lb/in.
K5 : K 6 : 20 k g f / m m : 196 k N / m : 1120 lb/in.
C~ = (72 : C4 : 1.2x10-z k g f s / m m : 0.118 m N s / m : 0.672 lb s/in.
TAKAYAMA AND YAMAGISHI ON TIRE VIBRATION 45

C3 : C5 = C6 : lxl0 ~ kgfs/mm = 0.098 m N s / m = 0.566 lb s/in.


D= 1.5x103 kgfs/mm = 14.7 M N s / m = 8.4x105 lb s/in.
r= 298mm=ll.73in.
W= 385 kgf = 3.776 kN = 849 lbf
h= 9 mm = 0.35 in.
Static Axial Forces

We now set all time derivatives in Eqs 3-7 equal to zero and choose C= 0.5 in
Eqs 9 to get En(O). We use this value in Eq 10 to get 60; equation 8 gives 61, Solution
of the five resulting simultaneous equations gives values for X1 X4 and ~:for any
series of 0 values. Substitution of these values in Eqs 13 and 14 gives the
corresponding values for Fr and Ft. The dotted lines in Fig. 9 show these
calculated values as functions of 0 through the entire footprint. These agree well
with the experimental values shown by the circles. Thus all unknowns are
established and the dynamic problem may be solved. The forces given in
kilograms may be multiplied by 9.8 to get the corresponding values in newtons or
by 2.2 to get pounds force.

"j Radial(Vertical)
i

Leading Edge Trailing-Edge

i[ ~.. TangentialJ

Leading Edge Trailing Edge

FIG. 9 - - Static axial forces.

Equations 13 and 14 may now be used to calculate radial and tangential forces
at any velocity. The dotted lines in Fig. 10, at V= 40 k m / h (25 mph), agree well
with the experimental values shown by the solid lines. The maximum oscillation
amplitudes FRpP and FTp~ may be used as indices of axial forces at a given speed.
Figure 11 shows these indices as a function of speed. Calculated values (dotted
lines) again agree well with experimental values (solid lines). The maximum
radial force occurs at 40 k m / h (25 mph) and the minimum in the region of 60-70
k m / h (37-44 mph). The tangential force peaks in the 30-50 k m / h (19-31 mph)
range.
46 TIRE SCIENCE TECHNOLOGY

:I ,-. [I Radial
~I'~ ['~ J~ (Vertical)

, o.o~"
sec. o . t

~ ~I~I ,~ Tangenl:~al
/'-~ (Horlzontal)

sec,

FIG. I0 -- Vibration of axialjbrces at tire speed of 40 km/h (25 mph).

o ~,B (Vertical)

tO0
>~(kmlh)

~ it Tangential

,90 I00

--~ ~ (kin~h)

FIG. 11 Maximum vibration amplitude o f axial forces as a function o f tire speed.

Resonant Frequency and Mode Analysis


Figure 12 shows that at 40 k m / h (25 mph) the torsional vibration at 40 Hz has
the lowest frequency of any vibrational mode. This particular resonance
frequency is strongly related to the tangential axial force, as discussed later.
The next lowest resonant frequency is that of the rigid mode of vibration in
either the tangential or radial direction, both of which are strongly related to
radial axle force. The difference between 67 Hz for the tangential and 74 Hz for
the radial directions may be caused by the different constraints. Resonance of the
TAKAYAMA AND YAMAGISHI ON TIRE VIBRATION 47

? ? ?
F I G . 12 -- Resonant frequency and vibration modes.

equivalent tread mass in rigid mode vibration would be at more than 100 Hz and
would be isolated from other motions. The physical meaning of this frequency,
however, is not clear.
Frequency Analysis
Plots of radial force as functions of frequency, such as those in Fig. 13, can
clarify the relative contributions to axial forces by vibrations at different
frequencies. Simulated results are seen to agree well with experimental results for
both radial and rigid mode vibrations.
The following conclusions may be drawn from the 40 k m / h data.
1. Radial axial force is almost dominated by the 75 Hz radial frequency.

Colculation Experiment
(3 I I

o Rodial 1 Rodiol
} ~--'A'(vertic~ } A (Verticol)

I o too 200
Frequency (Hz) Frequency (H~)

O~A'__ Tangential J -~ ~L/~L Tangential J


u_ 7 ](Horizontal)

T T
't I IVi
o too
A
20O
Frequency (Hz) To ,oo z~o
Frequency (Hz)
F I G . 13 - - Frequency analysis of axialJorcesfrom rigid mode o f vibration with tire rolling over
cleat at 40 k m / h (25 mph).
48 TIRE SCIENCE TECHNOLOGY

2. Tangential axial force is almost dominated by the 40 Hz torsional


frequency. A secondary contribution is made by the tangential rigid mode
vibration at 67 Hz.

Speed Dependency of Axial Forces


Figure 14 shows radial force vibrations at 40 k m / h (25 mph) and 80 k m / h (50
mph). In each case the vibrations cover 0.1 second after hitting a cleat. The
calculated traces (dotted lines) agree well with the experimental traces (solid
lines). The maximum amplitude at 40 k m / h occurs at the second peak and that at
80 k m / h occurs at the first peak. This indicates that transmissibility is the
dominant factor at the lower speed while input from the cleat and contact
pressures dominate at the higher speed. Therefore damping is more effective at
the low speed.

V = qOkrnlh ~-= 80kmlh


o

~'~ ':" l"t 2./F]-, Cglc,ul,~ ~8 ~Calculation-L-J

l It 111 I1 t! ~' ~t'l ~+'~- Y:I ~ M # 'k# ~ I~'l "T T~'~ i


IUI. II/ |81 P,fl-.Lq r J i~ ' I I
" l:I IX~-ExperimentJ ?~Experlment~
,~o
~ I,
o_o~
I ] ]2.~ NI,o i I I.,o .o6-I I I I Io.!. I
' t (sec.) ~ t (sec.)

FIG. 14 Build-up and decay of axial forces.

Simulated axial forces at 40 k m / h are separated in Fig. 15 into those caused by


rigid mode tangential vibrations and those caused by torsional mode vibrations.
The charts indicate that the two modes have nearly equivalent input to the
tangential forces but that the rigid mode dominates the radial (vertical) force.

Tangentail (Vertical) Radial (Vertical)

~F-I I r |r-~, Total ~L~ 8 ==L Rigid- Mode


~L I I 1.2' I . . . . j /~.1~
"~[]-]LT,~--I-L%L=]-RIgid Mode j 8

:4,t 2~
~Lt
~o
I i I o. I~-5" i t I t o . II o
~o
, o.o~" o.f
t (sec.) 9 t (secJ

FIG. 15 - Breakdown of axialJorces into rigid and tangential modes. Ordinates are kilograms
of force.
TAKAYAMA AND YAMAGISHI ON TIRE VIBRATION 49

Summary
Radial and tangential axial forces caused by a tire hitting a cleat on a test wheel
were obtained by using a mass-spring tire model having five degrees of freedom.
The rigid vibration mode and the torsional vibration mode were obtained by
taking into account the radial and tangential coupling effect. Simulated results
agreed well with experimental results in both the time domain and the frequency
domain. Radial rigid mode vibrations dominated the radial axial force, and
torsional vibration dominated the tangential axial force. Contribution of each
mode to the axial forces was clarified. More accurate results could be obtained by
taking into account the independent wheel torsional motion from the tire and
that of the tread shear spring at its contact with the ground.
References

[1] Majcher, J.S., "Simulation of Radial Ply Tires for Ride Tuning of Automobile Suspension",
Dissertation, University of Detroit, 1974.
[2] Yamagishi, K., and Jenkins, J.T., "The Circumferential Contact Problem for the Belt Radial
Tire" A SME Journal of Applied Mechanics, V ol. 47, 1980, pp. 513-518.
[3] Takebayashi, F., "On the Harshness of Compact Cars" Transactions ofJSA E No. 18, 1979,
pp. 103-109.
[4] Fiala, E., and Willumeit, H.P., "Radial Schwingungen von Gfirtel - - Radialreifen" A TZ, 68,
No. 2, 1966, pp. 33-38.
[5] Tielking, J.T., "Plane Vibration Characteristics of a Pneumatic Tire Model" SA E Paper No.
650492, 1965.
[6] B6hm, F., "Mechanik des Gdrtelreifens" A TZ 69, No. 8, 1967, pp. 255-261.
[7] Hirano, M. and Akasaka, T., "Natural Frequencies of the Bias Tire" Tire Science and
Technology, Vol. 4, No. 2, 1976 pp. 86-114.
[8] Akasaka, T. et al"Sixteenth Report on the Mechanics of Tires" (Vibration Characteristic of
Radial Tires), Technical Report B-22 of Chuo University, 1979, pp. 279-303.

You might also like