92

CHAPTER 4

SIMULATION AND ANALYSIS

OF STEERING KNUCKLE

4.1 INTRODUCTION

The performance study of product through fabrication and

testing is an expensive one in industry. Simulation and analysis of
components have been practiced in automobile industry, in order to reduce
the number of prototype during experimental testing. In this study,
computer aided design of steering knuckle was developed using modeling
package SOLID WORKS. Finite element analysis was performed on the
developed model of knuckle using HYPERMESH and ANSYS software to
simulate the effect under operating load conditions. This chapter discusses
the methodology adopted for static, fatigue, impact and modal analysis of
knuckle. Failure region identified from the anal ysis and optimized design
was proposed using Genetic Algorithm (GA) technique in this chapter to
achieve required fatigue life. The load-bearing behavior of optimized
design of steering knuckle was analyzed and reported for three materials:
MMC, SG iron and Al-Si alloy in a detailed manner in this chapter.

4.2 MCPHERSON SUSPENSION SYSTEM

In this study, steering knuckle used in McPherson suspension

system of a passenger car is considered in order to analyze loads at each
interface point of the knuckle. It combines a strut tube and a coil spring
 93

into a single unit, providing a more compact and lighter suspension system
more useful for a passenger car.

The loads acting on the steering knuckle connection point

during various riding conditions emanate from those generated at the tire-
road contact surface. The knuckle connecting the strut mount at the top,
lower arm at the bottom and tie rod on the side is shown in Figure 4.1.

Figure 4.1 Steering knuckle in McPherson suspension system

4.3 COMPUTER AIDED DESIGN AND ANALYSIS

Computer aided design and analysis consists of geometry

creation, applying material properties, meshing the component, defining
the boundary constraints and applying the proper load type to analyze the
load bearing behavior of any component.
 94

Steering Knuckle model

creation using Solid Works

Finite Element model

developed using
HYPERM ESH

Applying
Boundary Static Analysis using ANSYS Material Properties
Condition &
Multiaxial Load

Identifying Failure Region

Cyclic
Optimization
Material Fatigue Analysis using
using Genetic
Properties and ANSYS
Algorithm
Uniaxial Load
NO
Fatigue Life 1E+06 cycles
achieved

YES

Optimized Design

Impact Analysis

Modal Analysis

Results and Inference

Figure 4.2 Simulation and analysis procedure

The existing design and dimensions of a passenger car were

studied to develop the steering knuckle model in virtual environment.
Considering the geometry and load nature, the simulation procedure
 95

framed to study the behavior of knuckle under static and dynamic load is
shown in Figure 4.2.

4.3.1 Solid Modeling

The geometric model of the steering knuckle shown in

Figure 4.3 is developed using a modeling package SOLID WORKS
software Lombard (2007). The geometric data for the above model are
from M/s Sakthi Auto Components Ltd. India, supplying knuckle globally
for all types of vehicles. The dimension details are given in Appendix I.

Figure 4.3 CAD model of steering knuckle

The densities of MMC, Al-Si alloy and SG iron materials are

given as input in Solid works software and the corresponding weight
 96

properties are listed in Table 4.1. Rule of mixture was used to calculate the
density of MMC with varying weight percentage of TiC (Mallick 2007).

Table 4.1 Physical properties of steering knuckle

Density Weight
S.No. Materials
(kg/m3) (kg)
1 SGIron 7200 3.5
2 Al –Si 2700 1.0
3 MMC -10%TiC 2827 1.15
4 MMC -12%TiC 2854 1.2
5 MMC -15%TiC 2900 1.3

4.3.2 Meshing

The CAD model of the knuckle is imported in finite element

package Hypermesh for meshing process. It brings the virtual solid
component closer to the real solid component, excluding the manufacturing
defects. The selection of the elements was done on the merit of required
application. By considering the importance of mesh, knuckle model
divided into quality individual elements using Hypermesh software and
shown in Figure 4.4. The knuckle divided into 58, 783 elements and 99,046
nodes to minimize computational time. In this study a second order three
dimensional, 10-node element is used, due to its quadratic displacement
behavior and is well suited to modeling irregular meshes. The element is
defined by 10 nodes having three degrees of freedom at each node:
translations in the nodal x, y and z directions. The element has plasticity,
hyper elasticity, creep, stress stiffening, large deflection and large strain
capabilities.
 97

Figure 4.4 Finite element model of steering knuckle

4.3.3 Contact Surface and Boundary Condition

The developed finite element model is imported in ANSYS

software and contact was established at loading points to perform static
analysis. Since the steering knuckle has contact with four different
locations which is shown in Figure 4.5(a) the simulation was done using
contact points. The various components attached to the knuckle have been
replaced by pilot nodes as shown in Figure 4.5(b). The load acts on the
attached parts strut, tie rod, lower control arm and the wheel hub is
transferred to the required contact surface through five pilot nodes. Here,
node to surface contact is followed to transfer the load to the respective
region. TARGE170 element is used in to transmit the force from the pilot
node to the contact surface. To create contact between pilot node and
contact surface, CONTACT175 element is used in this analysis (Erdogan
 98

2011). Red color portion shown in Figure 4.5 indicate where the knuckle is
fixed with wheel hub.

Figure 4.5 Pilot node, contact surfaces and boundary conditions

Figure 4.6 Load transfer from pilot node to contact surface

 99

4.4 MULTIAXIAL LOAD CASES

The passenger car specifications for which the load cases are
analyzed are shown in Appendix II. The loads act on the steering knuckle
connection point during various riding conditions emanating from those
generated at the tire-road contact surface and var ying with vehicle
specification. Load cases for three condition namely cornering,
acceleration and braking were obtained from M/s Sakthi Auto Components
Pvt. Ltd. India, a leading manufacturer of knuckle components are utilized
in this work. These loads are to be applied to the finite element model
using the global vehicle coordinate system to get more reliable results.
The details of load cases are discussed below. The direction of forces
acting on the knuckle also vary in x, y and z directions at connection points
1, 2, 3 & 4 as shown in Figure 4.7.

Figure 4.7 Direction of loads in knuckle

 100

4.4.1 Cornering

During riding, when a vehicle taking a curved path of certain

radius of curvature about a centre is called cornering. This makes the
vehicle to transfer load in lateral direction from outer wheel to inner wheel
is shown in Figure 4.8(a). Lee (2001) proposed an equilibrium condition
(Equation (4.1)) while cornering a vehicle with a gravity force 1.2 g.
The force details for cornering load case are shown in Table 4.2 and its
direction and location shown in Figure 4.8(b).

(4.1)

Figure 4.8(a) Vehicle position in front view during cornering

 101

Table 4.2 Cornering loads

Location Force Moment

(N) (N-m)
Strut(1) -340 -270 -6750 -301920 78070 13430
Tie rod(2) -120 -830 140
Hub(3) -40 5260 6340 13764440 -17720 24110
Lower arm(4) 510 -4160 280

Figure 4.8(b) Multiaxial load directions during cornering

 102

4.4.2 Acceleration

While accelerating a vehicle, an inertia force acts on it at the

centre of gravity. As a result of this, a weight transfer takes place from the
front to the rear of the vehicle and tends to lift the vehicle as shown in
Figure 4.9(a). Considering a gravity force as 0.5g and the forces acting on
the vehicle along wheel base and the reaction forces will generate at front
and rear wheels to maintain equilibrium, which is shown in Equation (4.2)
Reza (2008).The details of acceleration load case given in Table 4.3 and its
direction are shown in Figure 4.9(b).

(4.2)

Figure 4.9(a) Vehicle position in side view during acceleration

 103

Table 4.3 Acceleration loads

Force Moment
Location
(N) (N-m) N-m
Strut(1) -340 -270 -6750 -301920 78070 13430
Tie rod(2) -120 -830 140
Hub(3) -40 5260 6340 13764440 -17720 24110
Lower arm(4) 510 -4160 280

Figure 4.9(b) Multiaxial load directions during acceleration

4.4.3 Braking

The load which is transferred from rear to front while applying

brake in a vehicle is called longitudinal load transfer. It tends to make the
 104

vehicle dive in forward direction as shown in Figure 4.10(a). Considering a

gravity force of 1.1g the equilibrium condition is given by Reza (2008)
shown in Equation (4.3). The braking load case details are given in
Table 4.4 and its position is shown in Figure 4.10(b).

(4.3)

Figure 4.10(a) Vehicle position in side view during braking

Table 4.4 Braking loads

Force Moment
Location
(N) (N-m) N-m
Strut(1) -340 -270 -6750 -301920 78070 13430
Tie rod (2) -120 -830 140
Hub (3) -40 5260 6340 13764440 -17720 24110
Lower arm (4) 510 -4160 280
 105

Figure 4.10(b) Multiaxial load directions during braking

4.4.4 Observations from Multiaxial Load Cases

It is observed from the load cases listed in Table 4.5 the

magnitude of the load acting on the lower arm joint of the knuckle during
braking is 8450 N is the highest one. This is due to vehicle load transfer in
longitudinal direction from rear side of the vehicle to front side and there
by knuckle joint experiences more load. During cornering case, the axis of
rotation of wheel in the longitudinal direction gets changed through a small
angle which induces gyroscopic effect. Therefore a maximum of 13764440
N-mm moment is acting at hub during cornering. In each load case
12 force components and six moments contribute load simultaneously on
the knuckle in different directions. It is understood from the study, both
point load and moment loads act at strut and hub region of the knuckle for
 106

all the load cases. However at steering arm and lower arm joints, only force
components acts on the knuckle. Also it was observed from the loads
during braking, magnitude of force and moments were larger at all the
connection points when compared to cornering and acceleration. The
direction of loads varies continuously with respect to driving condition,
and vehicle specification.

Table 4.5 Multiaxial loads summary

X Y Z X Y Z
Location Load cases
N N-mm
0.8g cornering -340 -270 -6750 -301920 78070 13430
Strut
0.5 acceleration 70 -1080 3140 287110 135700 -40560
joint
1.1g braking 1620 -1690 -4940 388030 796340 -146560
0.8g cornering -120 -830 140
Steering
0.5 acceleration 130 1030 90
arm
1.1g braking 60 560 -80
0.8g cornering -40 5260 6340 13764440 -17720 24110
Hub 0.5 acceleration -1690 20 2960 3580 0 10
1.1g braking 6780 -140 5670 -41600 -1767550 -27650
0.8g cornering 510 -4160 280
Lower
0.5 acceleration 1490 30 80
arm
1.1g braking -8450 1270 -650

4.5 STATIC ANALYSIS

Steering knuckle breakage mainly occurs due to multiaxial load

condition is encountered in service. Hence, determining the failure location
exactly becomes important for design optimization. Static analysis using
 107

ANSYS software provides information about stress concentration region

for the applied load and material condition. The component behavior and
material properties are assumed to be linearly elastic in static analysis.
It determines the displacements and vonmises stresses in knuckle caused
by loads that do not induce significant inertia and damping effects. Steady
loading and response conditions are assumed in this analysis where the
loads and the structure response are assumed to vary slowly with respect to
time. The material properties used for the analysis are shown in
Table 4.6. The analysis provides details about the stress pattern, stress
concentration region and maximum deflection, of the knuckle for the
applied loads.

Table 4.6 Material properties

Modulus of Poisson’s
S.No. Materials
Elasticity(E) ratio(m)

1 SG iron 170 0.28

2 Al-Si alloy 70 0.3

2 MMC -10 wt.% TiC 79 0.3
3 MMC -12 wt.% TiC 84 0.3
4 MMC -15 wt.% TiC 87 0.3

4.5.1 Static Analysis Results

The effect of multiaxial loads on knuckle for the above said load
cases were compared through vonmises stress and deflection plot.
Simulation of knuckle under cornering, acceleration and braking load cases
are compared in Figure 4.11(a). Since the static analysis is performed
 108

within elastic limit, the observed vonmises stress is 26.53 MPa during
cornering.

90
78.79
Vonmises stress, MPa

70 Cornering

Acceleration
50
34.8 Braking

30
26.53

10 12% TiC

15% TiC
10% TiC
SG
Al-Si

Materials Considered

Figure 4.11(a) Vonmises stress vs materials

0.07
Cornering

0.06 Acceleration

Braking
0.05 0.049
Deflection , mm

0.045
0.043 0.042 0.041
0.04 0.04
0.039 0.037 0.036
0.036
0.03 0.034 0.033

0.021
0.02
0.018
0.017
0.01

0
10% TiC

12% TiC

15% TiC
Al-Si

SG

Materials considered

Figure 4.11(b) Deflection vs materials

 109

(a)

Figure 4.12(a) Vonmises stress - cornering

Figure 4.12(b) Vonmises stress - acceleration

 110

Figure 4.12(c) Vonmises stress - braking

Figure 4.13(a) 15% TiC MMC deflection - cornering

 111

Figure 4.13(b) SG iron deflection - cornering

Figure 4.13(c) Al-Si alloy deflection - cornering

Though the stress value increase to 34.8 MPa and 78.79 MPa for
acceleration and braking condition when the stress is constant for all the
materials. The variation observed with respect to different load case infers
 112

that braking condition creates more stress in the knuckle. The obtained
stress values were compared with yield strength of the respective materi al
shows that all the stresses are within the elastic limit. However, stress
concentration is more at steering arm region in all cases. It is observed
from the Figure 4.11(b), deflection plot of SG iron knuckle varies from
0.021 mm to 0.045for Al-Si alloy 0.039 to 0.049 for MMC 0.033 to 0.045.
Also it is seen that the deflection 0.05 mm is the highest one occur to Al-
Si alloy in cornering load case compared to other materials.
The deflection values obtained from the analysis shown in Figure 4.11(b)
varies between 0.01 to 0.06 mm which is very negligible. The results show
that the knuckle structure was not affected greatly by static load condition.
The vonmises stress contour plot for three load cases is shown in
Figure 4.12(a-c) and it clearly indicates t hat the stress concentration pattern
repeatedly occur on steering arm region for all the materials considered.
The deflection contour plot for three load cases shown in Figure 4.13(a-c)
pointed that the deflection is more at the steering arm end in red colour
region. All the analysis results reported in this section assumed that the
load is not varying with respect to time. But in real case the load is cyclic
and varying with time. Therefore to study the behavior of knuckle under
cyclic load, fatigue analysis is performed in the next section.

4.6 FATIGUE ANALYSIS

The performance of the component at design stage is to perform

the fatigue life analysis after static analysis and optimized the design with
respect to results derived from static analysis. The fatigue properties of a
structural component represent its response to cyclic loading, which is a
common occurrence in automobile. In practice, loads significantly below
static limits can cause failure if the load is repeated sufficient times
characterizing the capability of a material to survive under cyclic loads.
 113

The strength of a material is significantly reduced under cyclic loads and

failed in a brittle manner when they are subjected to repeated cyclic
stresses. It usually involves the initiation and growth of a crack until it
reaches a critical size, sometimes causing separation into two or more
parts. Fatigue failure of a component depends on stress level, stress state,
mode of cycling, process history, material composition, and environmental
conditions. The time history of stress or strain, at the exact location where
a crack is going to start, is the critical factor which is determined by
conventional experiments. It involves a lot of time and cost to identify the
critical zone and life of the component.

Fatigue analysis has traditionally been performed at a later stage

of the design cycle. This is due to the fact that the loading information
could only be derived from the direct measurement, which requires
experimental results. The purpose of analyzing a structure early in the
design cycle is to reduce the development time and cost. This is achieved
to determine the critical region of the structure and improve the design
before prototype testing. The finite element based fatigue analysis can be
considered as a complete engineering analysis for the component subjected
to time varying loads. By using this analysis it could be easy to determine
critical location within a model and concentrate more on it. Moreover, it
was observed from the results of static analysis of a steering knuckle for
multiaxial load condition of steering arm region is susceptible for fatigue
failure under cyclic load condition. Therefore passenger car operating load
of 3.6 kN was applied on that region of the knuckle for 1.0+E06 cycles to
determine the life and damage of the knuckle. The details of fatigue
analysis procedure are discussed in the following section.
 114

4.6.1 Fatigue Analysis Procedure

Fatigue techniques were classified based on two methods such

as stress-life method and strain-life method. Strain-life method was
generally used to characterize low-cycle fatigue life of the components.
Strain-life is typically concerned with crack initiation with a relatively low
number of cycles which is lesser than 10 5 cycles. Stress-life approach is
with total life and does not distinguish between initiation and propagation.
In terms of cycles, stress-life typically deals with a relatively high number
of cycles which is more than 10 5 cycles inclusive of infinite life.
Since steering knuckles subjected to high number of cyclic loads during its
function, stress-life approach is used in this study to conduct fatigue
analysis. In this approach S-N curve of the material was given as input in
the ANSYS software material database. The S-N curve details of five
material classifications (SG iron, Al alloy, M MC with 10, 12 and 15%
MMC) are used to analyze the fatigue performance of the knuckle.
The values to plot S-N curve for MMC, SG iron and Al-Si alloy was taken
from the experimental results discussed in Section 3.8.3 and from the
literature of ASM (1978). Fatigue analysis is carried out in ANSYS
software using stress- life approach to determine the fatigue life and
damage of the knuckle due to cyclic load before and after optimization of
knuckle. It has been observed from the steering knuckle manufacturing
industry, the knuckle should withstand 1+E06 cycles during experimental
test for the applied load without any crack. Therefore in this study a design
life of 1+E06 cycles was considered as minimum target to analyze the
knuckle. The load applied on the knuc kle at steering arm end is shown in
Figure 4.14.
 115

3.6 KN

Figure 4.14 Fatigue analysis at steering arm region

4.6.2 Loading Type

Load range is the most important parameter in fatigue analysis

and it is defined by the amplitude and mid value. Practically the load
histories of automobile components are complex in nature as shown in
Figure 4.15(a). Therefore load type is assumed to be fully reversed with
constant amplitude as shown in Figure 4.15(b). The loading ratio is defined
as the ratio of the maximum value to minimum value. Constant type fully
reversed load R = -1 is applied at the steering arm end which is equal in
magnitude and opposite in direction as given below in Equation (4.4).

Load ratio= Max/Min= 3.5/-3.5= -1 (4.4)

 116

Figure 4.15 (a) Real time load history (b) Constant amplitude fully
reversed load

4.6.3 Mean Stress Correction

In completely reversed constant amplitude tests the testing of

cyclic fatigue properties were carried out in which the mean stress is zero
as shown in Figure 4.15(b). But in real case the loading is not completely
reversed and mean stress exists. To consider this effect mean stress
correction is included in the analysis. Generally there are three methods
available for mean stress correction: Soderberg, Goodman and Gerber
methods. Considering the ductile nature of material properties Soderberg
method was considered in this study to include the mean stress effect
during analysis (Dieter 1988).

4.6.4 Fatigue Analysis Results and Discussion

The fatigue life of the MMC steering knuckle calculated based

on stress- life approach is discussed in this section. The contour plot results
shown in Figure 4.16 (a- c) shows that the fatigue life is 1.65 E+06 cycles
 117

for 15 % TiC, 12.8 E+06 cycles for 12 % TiC and 9.89 E+06 cycles for
10% TiC. All the above mentioned life cycles is on the higher side when
compared to SG iron (8 E+05 cycles) and Al-Si alloy (3.97 E+05 cycles)
respectively (Figure 4.16 (d) & (e)). This is attributed to better fatigue
properties of MMC compared to conventional material.

In fatigue analysis damage factor indicates the ratio of applied

target cycles to the total life of the knuckle. The applied cycles of 1E+06
cycles are used in this study. Therefore, the damage less than one indicates
that the knuckle survived the required fatigue life of 1E+06 cycles without
any physical damage for the applied load of 3.6 kN. It was observed from
the analysis that the damage value less than one for 12%TiC (0.7) and 15%
TiC (0.6) reinforcement knuckle. The damage contour plot for Al-15 wt.%
TiC MMC knuckle is shown in Figure 4.17(a) which indicates that the
damage of the knuckle during cyclic load started where the stress
concentration is more. The damage red spot in enlarged view shown by
arrow mark is the starting point for crack initiation. In the case of SG iron
knuckle, Figure 4.17(b) shows that the value is 1.25 and circular red
damaged area is almost similar to MMC knuckle. But for aluminum alloy
the damaged red spot is active and shown by elongated red line in
Figure 4.17(c) with a value of 2.5. In all the cases, crack starts its path from
the steering arm edge where stress concentration is more. With the existing
knuckle design MMC with 10wt% TiC, SG iron and Al-Si alloy were all
failed due to this damage and could not achieve its design life of 1+E06
cycles.
 118

Figure 4.16(a) Fatigue life of 15% TiC reinforced MMC

Figure 4.16(b) Fatigue life of 12% TiC reinforced MMC

 119

Figure 4.16(c) Fatigue life of 10% TiC reinforced MMC

Figure 4.16(d) Fatigue life of SG iron knuckle

 120

Figure 4.16(e) Fatigue life of Al-Si alloy knuckle

Figure 4.17(a) Damage spot of 15% TiC reinforced MMC

 121

Figure 4.17(b) Damage spot of SG iron

Figure 4.17(c) Damage spot of Al-Si alloy

4.7 DESIGN OPTIMIZATION OF KNUCKLE

Optimization is a process of finding maximum or minimum

value of a function which provides the best solution under given
 122

circumstances. The static and fatigue simulation discussed in previous

section shows that steering arm region of the knuckle is critical under
operation condition. Hence, while going for light weight material the
existing cross section in this region will lead to failure. Therefore in order
to minimize the stress and deflection at steering arm region a design
optimization was carried out under functional and geometric constraints.
Conventional optimization technique operates on mathematical function
which is continuous and differentiable to provide optimum solution. Since,
the objective and constraint functions are separable in the design variables,
the conventional technique is not suitable in the present work. Therefore
non-conventional technique genetic algorithm is selected to obtain
optimized solution. It is observed from the load and boundary condition of
knuckle, steering arm region is subjected to bending and exhibit stress
concentration as shown in Figure 4.18(a). Therefore to alter the cross
section shown in Figure 4.18(b) where maximum stress occurs, steering
arm has been considered as cantilever beam as shown in Figure 4.18(c).

Failure Region

Figure 4.18(a) Optimization region in steering knuckle

 123

Figure 4.18(b) Steering arm existing cross section at failure region

before optimization

Figure 4.18(c) Steering arm region approximate as cantilever beam

4.7.1 Optimization Problem Formulation

The purpose of formulation for a physical problem is to create a

mathematical model of optimum design, which can be solved using an
optimization technique. The objective function to minimize the stress and
deflection of a beam for a load of 3.6 kN at critical cross section is given in
Equations (4.5) and (4.6). In the present problem cross section of critical
 124

regions were considered as design variables which are shown in

Equations (4.7) and (4.8). The maximum and minimum limits are given in
Equations (4.9) and (4.10). The constraint equations control the solution by
considering the design requirements. The present problem has two
constraints. Firstly the beam weight is not more than 20% of the existing
one after optimization. It depends on dimension of existing cross section,
density and length. The second one is: moment of inertia of the optimized
beam is always greater than the present one. These two constrains are
shown in Equations (4.11) and (4.12). The parameters which affect the
objective function are known as design parameters. Modulus of elastic and
density of MMC and that of SG one given as input during optimization

 Objective Function

(4.5)

(4.6)

 Design Variables

(4.7)

(4.8)

 Maximum and Minimum Limits

(4.9)

(4.10)
 125

 Design Constraints

(4.11)

- 80 mm; =23 mm;

(4.12)

-area moment of inertia,mm4

4.7.2 Genetic Algorithm

Genetic algorithms are based on the principles of natural

genetics and natural selection. It is used for solving both constrained and
unconstrained optimization problems with near global optimum solution.
GA repeatedly modifies a population of individual solutions till, the
population "evolves" toward an optimal solution based on the genetic
principle “survival of the fittest”. Kalyanmoy (1995) reported that genetic
algorithm uses three main types of rules at each step to create the next
generation from the current population. Selection rules select the
individuals, called parents that contribute to the population at the next
generation. Crossover rules combine two parents to form children for the
next generation. Mutation rules apply random changes to individual parents
to form children.

Fitness function is arrive d from the objective function

formulated in previous section which is used in successive generations.
The best fitness value gives the optimized dimensions for the given
constrained function. The fitness function for the present study is given in
Equation (4.13). The fitness value gives the optimized dimensions for the
given constrained function.
 126

h( = { f( ,g( } (4.13)

The parameters control the time taken and the accuracy of

solution. Therefore setting the parameters is an important stage in
optimization to achieve optimal solution. In GA, the design variables are
represented as strings of binary numbers, 0 and 1. It represents the
individuals generated randomly in the initial population. Rank method
chosen as selection function in this study, which selects the individual as
parents for the next generation depends on the rank of the score of the
fitness function.

Table 4.7 Reproduction parameter

S.No. Parameters Nos.

1 Initial Population 20

2 Elite count 2

3 Cross over 16

4 Mutation 4

5 Maximum generations 100

Reproduction parameters for the present study are shown in

Table 4.7. For the reproduction, the elite count is taken about 20% of the
population, 80% of the remaining individual undergoes crossover.
After performing trial runs, the maximum number of generation is fixed as
100. The flowchart shown in Figure 4.19 describes the procedure of
optimization.
 127

Figure 4.19 Optimization procedure using genetic algorithm

 128

4.7.3 Optimization Results and Discussion

The obtained best individuals after completing 100th generation

is shown in Figure 4.20 (a-b) represent the optimized cross section of
MMC and SG iron. Figure 4.20 (c-d) gives the minimum value of
deflection for the optimized cross section. Optimized dimensions of
cantilever beam obtained after optimization shown in Table 4.8. It is
observed from the table, bending stress is 125 N/mm2 for all the materials
considered. The objective function shown in Equation (4.5) not having any
variable to control the material properties is the reason for same stress
value for all the materials. Figure 4.20(e) shows bending stress plot of
MMC beam. However the deflection is 0.7 mm for SG iron is the highest
value compared to MMC having 0.5.
Dimensions, mm

Design variables,

Figure 4.20(a) Optimized cross section of MMC Beam

 129

Dimensions, mm

Design variables,

Figure 4.20(b) Optimized cross section of SG iron beam

Deflection, mm

Generated population

Generated population

Figure 4.20(c) Deflection of MMC beam

 130

Deflection, mm

Generated population

Figure 4.20(d) Deflection of SG iron beam

2
Bending Stress, N/mm

Generated population

Figure 4.20(e) Bending stress of MMC beam

 131

Objective function shown in Equation (4.6) has modulus of

elasticity. Therefore the deflection of beam varies for MMC and SG iron.
Optimized cross section (23 X 27) shown in Figure 4.20(f) is the required
cross section to be implemented at critical region of the MMC knuckle to
withstand operating load of 3.6 kN. It is concluded from the results, by
changing the dimensions of the critical region, the knuckle design can be
improved to withstand the operating load. This design modification is
verified in next section through fatigue analysis.

Table 4.8 Failure region cross section details after optimization

SG MMC
Optimized Parameters Al-Si Alloy
Iron 15% TiC

Maximum Bending Stress, N/mm2 125 125 125

Maximum Deflection, mm 0.56 0.7 0.51
Cross section Dimensions, mm 25 X 27 22X23 23X27

Figure 4.20(f) Optimized cross section at failure region

 132

4.8 FATIGUE ANALYSIS RESULTS AFTER

OPTIMIZATION

After optimization the knuckle was modified at failure region

and had undergone fatigue analysis to evaluate its performance unde r
uniaxial load condition of 3.6 kN. It was observed from the Figure 4.21(a),
after optimization life period of the MMC knuckles varied from 1.16E+06
cycles to 1.94E+06 cycles. It was evident from the comparison shown in
Figure 4.21(a), failure region optimization effectively increases the fatigue
performance of the knuckle greatly for the materials considered.
Also MMC knuckle with 15% TiC can survive a minimum 1.948 E + 07
cycle which is shown in Figure 4.21(b). Fatigue life of SG iron knuckle
withstands 1.07E + 06cycles is shown in Figure 4.21(c). It is observed
from the comparison of 15 wt.% TiC knuckle with SG iron MMC has 82%
more life. It is seen from the Figure 4.22(a) the damage value was reduced
after optimization.

After Optimization
Before Optimization
4.00.E+06

3.50.E+06 1.95E+06

3.00.E+06
1.52E+06
No of Cycles

2.50.E+06
1.16E+06
2.00.E+06
1.07E+06 1.65E+06
1.50.E+06
8.43E+05 1.29E+06
1.00.E+06 9.89E+05
8.00E+05
5.00.E+05
3.97.E+05
0.00.E+00
Unreinforced SG Iron Al-10%Tic Al-12%Tic Al-15%Tic
Alloy
Materials

Figure 4.21(a) Fatigue life comparison before and after optimization

 133

Figure 4.21(b) Fatigue Life of 15% TiC MMC after optimization

Figure 4.21(c) Fatigue life of SG iron after optimization

 134

3
Damage before Optimization
2.5
2.5 Damage after Optimization

2
Damage

1.5 1.25
1
1 0.7
1.18
0.9 0.6
0.8
0.5 0.6 0.5

0
Unreinforced Alloy SG Iron Al-10%Tic Al-12%Tic Al-15%Tic
Materials

Figure 4.22(a) Damage comparisons before and after optimization

Figure 4.22(b) Damage of 15% TiC reinforced MMC after

optimization
 135

Figure 4.22(c) Damage of SG iron after optimization

Damage value of optimized model shown in

Figure 4.22(a) shows the damage is reduced from 0.6 to 0.5 due to
optimization. Figure 4.22 (b & c) shows crack initiation point occurs on the
surface whereas before optimization it started from the edge. It is
understood from the results, the fatigue life of the knuckle has been
increased due to optimization.

4.9 IMPACT ANALYSIS OF STEERING KNUCKLE

Impact strength is one of the greatest design limitations involved

in designing new products. The purpose of impact analysis performed in
this work is to gain an initial understanding of the impact behavior of
steering knuckle during sudden shock load in a very short period of time.
Impact properties such as damage resistance are crucial to product design
due to the significant deterioration of the material properties after impact
damage has occurred. The impact force applied on the knuckle steering
arm region is shown in Figure 4.23. The force required for impact is
calculated using Equation (4.9), where m is the mass of the projectile, v the
 136

impact velocity, and t the time duration. The impact duration considered in
this study is about 0.001 second and the mass is 87.3 kg. The impact force
is applied to nodes at a certain area, which is characterized by the shape of
the projectile. An impact force that will be equivalent to a projectile with
an impact velocity of 2, 3 and 4 m/s was calculated using Equation (4.14)
and it is applied on the nodes of the impact location. The maximum values
of deflection which vary with the values of the impact forces are discussed.
To perform impact analysis explicit finite element solver RADIOSS was
used in this study.

(4.14)

Figure 4.23 Steering arm impact test

 137

4.9.1 Impact Results And Discussion

It was observed from the Figure 4.24 (a-c) for the velocity of
2 m/s the deflection varies between 3.5 mm to 4.3 mm as shown at the free
end. The steering arm region affected by an impact force of 17.5 kN shown
in Figure 4.24(a) is 3.5 mm for 15%TiC reinforced MMC. The blue region
of the knuckle contour plot is large in this case which showed that the
remaining portion of the knuckle was not affected largely. However, it was
observed from Figure 4.25 (a-c) for 3 m/s and Figure 4.26 (a-c) 4 m/s the
impact force is 26.2 kN and 35 kN which affects the knuckle structure
severely. The deflection 5.19 mm at steering arm end for 3 m/s is shown in
Figure 4.25(a) for 15% TiC.

Figure 4.24(a) 15 wt.% TiC MMC knuckle at 2 m/s

 138

Figure 4.24(b) SG iron knuckle at 2 m/s

For 4 m/s 8.6 mm is the deflection which is shown in

Figure 4.26(a). By comparing the contour plot at each velocity, the entire
portion of the knuckle is affected largely. This can be seen by the reduction
of blue region when the velocity is increased from 2 to 4 m/s. For all the
three velocity cases the deflection value is compared in Figure 4.27 which
shows that the MMC steering knuckle exhibits better values compared to
SG iron and Al-Si alloy. Therefore for experimental analysis 4 m/s could
be used as impact velocity to study the performance of knuckle under
impact load.
 139

Figure 4.24(c) Al-Si alloy knuckle at 2 m/s

Figure 4.25(a) 15 wt.% TiC MMC knuckle at 3 m/s

 140

Figure 4.25(b) SG iron knuckle at 3 m/s

Figure 4.25(c) Al-Si alloy knuckle at 3 m/s

 141

Figure 4.26(a) 15 wt.% TiC MMC knuckle at 4 m/s

Figure 4.26(b) SG iron knuckle at 4 m/s

 142

Figure 4.26(c) Al-Si alloy knuckle at 4 m/s

14
12.4 4m/s

12
10.3 3m/s
10 9.3 9.7
Deflection

8.6
mm

2m/s
8 6.9
5.5 6.3
5.9
6 5.19

4
4.33
3.59 3.7 3.8 3.9
2

0
15% 12% 10% SG Al alloy

Materials considered

Figure 4.27 Deflection comparison of materials considered

 143

4.10 MODAL ANALYSIS

Modal Analysis is an important technique in detecting and

interpreting natural frequencies of a structural component (Rao 2009).
The natural frequency of any material is influenced by elastic constant,
density and geometrical dimensions as reported in literature (Lai 2011).
Vibratory conditions produce stresses, which exceed allowable fatigue
strength and in some cases even result in failure. Therefore prior
assessment of these responses followed by corresponding corrective
actions ensures safe design. It is currently used by several automotive
industries to study the dynamic character of a component with respect to
loads applied to the system. The steering knuckle location in an automobile
is critical and directly experiences the vibration from the road undulation
during vehicle motion. Therefore the study of modal characteristics is
significant in this study to avoid resonance excitation during its function.
The material properties and density are given as input parameters in
ANSYS software to study the performance of knuckle under free vibration
condition. The natural frequency and mode shapes for the first 10 modes of
the knuckle obtained from modal analysis are discussed below.

4.10.1 Results and Discussion

Since the analysis type is free suspended vibration the first six
modes are rigid modes and does not influence the vibration characteristics.
The operating frequency of a passenger car varies between 1500 to
3000 Hz. Therefore the natural frequency obtained from seventh mode
plays a significant role in resonance excitation. The natural frequency for
the materials considered is shown in Table 4.9.
 144

Table 4.9 Natural frequency comparison

Materials Frequency (Hz)

Modes 1-3 4 5 6 7 8 9 10
SG iron 0 7.80E-03 1.2E-02 1.60E-02 1675 1806 2522 2817
15% TiC 0 8.50E-03 1.10E-02 1.80E-02 1767 1899 2658 2967
12% TiC 0 8.80E-03 1.20E-02 1.80E-03 1798 1932 2704 3019
10% TiC 0 8.90E-03 1.20E-02 1.80E-02 1836 1974 2763 3084
Al-Si alloy 0 9.00E-03 1.20E-01 1.90E-02 1861 2000 2799 3124

3300

3100

2900

2700
Frequency (Hz)

2500

2300

2100

1900

1700

1500
SG iron 15% TiC 12% TiC 10% TiC Al alloy
Matrerials

Figure 4.28 Steering knuckle natural frequency

For existing SG iron knuckle the natural frequency varies from

1675 to 2817 Hz. The frequency range gradually increases for MMC
knuckle when the weight percentage of TiC is decreased. Among the
considered materials Al-Si alloy natural frequency is maximum due to low
 145

elastic constant. It was observed from the Figure 4.28for the seventh mode
the difference between SG iron and MMC is 131 Hz for 10% TiC and
92 Hz for 15% TiC.

7th mode 8th mode

9th mode 10th mode

Figure 4.29 Mode shapes of 15% TiC MMC knuckle

Similarly the trend follows for the remaining modes 8, 9 and 10

also. The frequency spectrum variation above 500 Hz only can suspect for
resonance excitation. Therefore the MMC steering knuckle could be safely
work in the steering system with the available frequency spectrum.
The finite element predicted mode shapes for three materials is shown in
Figure 4.29.
 146

4.11 CONCLUSIONS

 The solid model of the passenger car steering knuckle is created

using modeling software SOLID WORKS

 Finite element modeling was developed using HYPERMESH

software

 Multiaxial load analysis conducted for three load cases and the
results were compared for the MMC, SG iron and Al-Si alloy.
The results indicate that the steering arm region of knuckle is
critical

 Fatigue analysis conducted on steering arm region using stress-

life approach. The damage and design life of the knuckle was
found using fatigue analysis for a design life of 1E+06 cycles

 Genetic Algorithm technique was used in Matlab software for

optimization of failure region. Uniaxial load value of 3.5kN was
applied at steering arm region and optimized dimension
(23X27) were implemented in the critical cross section region of
knuckle

 Fatigue life and damage of knuckle were compared before and

after optimization of knuckle

 Impact analysis were performed on the knuckle at 2, 3 and 4 m/s

to study the plastic deformation due to impact load

 Modal analysis also conducted on the knuckle to determine

mode shape and natural frequency of the MMC
 147

 From all the results and inferences obtained from finite element
analysis and optimization process, it can be observed the
steering knuckle could be safely replaced SG iron with
improved performance