DESIGN OF A CARBON FIBRE PASSENGER CAR RIM

Lehoelere Frederick Maserumule

A research report submitted to the Faculty of Engineering and Built Environment of

the University of the Witwatersrand in partial fulfilment of the requirements for the
degree of Master of Science in Engineering.

Johannesburg 2015
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Declaration

I Lehoelere Frederick Maserumule hereby declare that this research report is my

own unaided work. It is being submitted to the Degree of a Masters in Science in
Engineering to the University of the Witwatersrand, Johannesburg. It has not been
submitted before for any degree or examination to any other University.

……………………………………………………………………………
L.F Maserumule 0505585E

……….. day of …………….., ……………

ii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Dedication

In memory of my late aunt Masemola Granny

iii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Acknowledgements

I would like to acknowledge Dr Frank Kienhofer (my supervisor) for his assistance in
obtaining the strain gauge experimental test results of the 1098 Ducati Rim. His
supervisory role provided the necessary assistance and guidance in the key aspects
of this research report.

The guidance of Craig Goodrum was also invaluable in understanding composite

theory, and the manufacturing process of the rim at BlackStone Tek.

I would like to thank Makgomo Maserumule, Frank Maserumule, Tseke Maserumule,

Mamagane Maserumule and Ayanda Mkhasibe for the emotional support during the
research process.

iv
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Abstract

The wheel is an important component for all land transport vehicles and the following
research report investigates the feasibility of a passenger car wheel manufactured
from carbon fibre. Previous research, evaluated a strain gauged BlackStone Tek
motorcycle Ducati rim loaded to 700 Nm on a dynamic cornering test machine. In the
current study, these experimental results were used to validate an FEA model of the
BlackStone Tek Ducati rim. The validated FEA was used to investigate the feasibility
of using carbon fibre to manufacture a passenger car rim. The FEA model did not
include the aluminium hub, and the analysis was therefore considered valid for the
spokes and rim only. The FEA strains of the Ducati rim spokes were within an
acceptable 25% correlation range with the experimental results. The FEA strains
were generally higher than the experimental results. Having validated the FEA
model, a carbon fibre passenger car rim, was analysed at a bending moment of 2000
Nm. The inverse of the factor of safety from the spoke to the rim section was below
1. This indicates that the newly designed rim should pass the dynamic cornering test,
a vital standard proving that the rim is structurally durable to function on a small to
medium sized passenger car. In comparison to aluminium rims of the equivalent
size, a weight saving of 62.2% can be expected from a carbon fibre passenger car
rim.

v
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Table of Contents

Declaration .......................................................................................................................... ii

Dedication .......................................................................................................................... iii

Acknowledgements ........................................................................................................... iv

Abstract ............................................................................................................................... v

Table of Contents ............................................................................................................... vi

List of Figures .................................................................................................................. viii

List of Tables ...................................................................................................................... xi

List of Symbols ................................................................................................................. xii

Chapter 1: Introduction ...................................................................................................... 1

1.1 Background ........................................................................................................................... 1

1.2 BlackStone Tek Ducati Rim ................................................................................................ 3

1.3 Wheel Test Methods ............................................................................................................ 4

1.4 Problem Description............................................................................................................. 6

1.5 The Proposal ......................................................................................................................... 7

Chapter 2: General Theory ................................................................................................. 8

2.1 Design Standards for Wheels ............................................................................................. 9

2.2 Experimental Analysis of the Ducati Rim........................................................................ 12

2.2.1 Stress strain relationship of a lamina ...................................................................... 12

2.2.2 Principles of strain gauging ....................................................................................... 14
Chapter 3: Literature Review ............................................................................................ 16

3.1 Experimental Analysis ....................................................................................................... 16

3.2 Numerical Analysis............................................................................................................. 19

3.3 Previous Work on Composites ......................................................................................... 21

Chapter 4: Objectives ....................................................................................................... 22

4.1 Research Scope ................................................................................................................. 22

4.2 Research Outline ................................................................................................................ 23

vi
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Chapter 5: Experimental Analysis: 1098 Ducati Rim ...................................................... 24

Chapter 6: FEA: 1098 Ducati Rim..................................................................................... 27

6.1 Methodology ........................................................................................................................ 27

6.1.1 Model preparation ...................................................................................................... 27

6.1.2 Project schematic ....................................................................................................... 29
6.1.3 Material design ........................................................................................................... 30
6.1.4 Meshing ....................................................................................................................... 39
6.1.5 Load cases .................................................................................................................. 41
6.2 Result Evaluation: Current 1098 Ducati Rim ............................................................... 43

6.2.1 Strains .......................................................................................................................... 43

6.2.2 Maximum strain: failure criterion .............................................................................. 48
Chapter 7: FEA – Passenger Car Rim .............................................................................. 56

7.1 Ducati Rim Exposed to Passenger Vehicle Loads........................................................ 58

7.1.1 Strains .................................................................................................................................. 58

7.1.2 Maximum strain failure criterion ....................................................................................... 61

7.2 Model Preparation .............................................................................................................. 64

7.2.1 Segmentation .............................................................................................................. 65

7.2.2 Stacking sequence ..................................................................................................... 65
7.2.3 Meshing ....................................................................................................................... 71
7.3 Result Evaluation: Passenger Vehicle Rim.................................................................. 72

7.3.1 Strains .......................................................................................................................... 72

7.3.2 Maximum strain failure criterion ............................................................................... 77
Chapter 8: Discussion and Conclusion ........................................................................... 83

8.1 Discussion ........................................................................................................................... 83

8.2 Conclusion and Recommendations................................................................................. 88

References ........................................................................................................................ 89

Appendix A ........................................................................................................................ 94

Micromechanics of composite material ................................................................................... 95

vii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

List of Figures

Figure 1: 1098 Ducati BlackStone Tek Rims [5]. ........................................................ 1

Figure 2: Anatomy of the BST Ducati Rim ................................................................. 4
Figure 3: Wheel Test Methods [10; 11] ..................................................................... 5
Figure 4: Wheel Dynamic Variables [23] .................................................................... 8
Figure 5: 1098 Ducati Rim Setup for Cornering Fatigue Test ................................... 10
Figure 6: Unidirectional Fibres [28]........................................................................... 12
Figure 7: Strain Gauge Rosette [29] and Wheatstone bridge Circuit [2] ................... 14
Figure 8: 45 Degree Rosettes Instrumented on the Wheel [edited 11] ..................... 17
Figure 9: Strain Gauged Ducati Rim and Experimental Setup [edited 8] .................. 24
Figure 10: Strain Gauge Results of 1098 Ducati Rim during One Test Run [8] ........ 25
Figure 11: Surface Model of the 1098 Ducati Rim .................................................... 27
Figure 12: Illustration of 1098 Ducati Rim Inner and Outer Sections ........................ 28
Figure 13: Project Schematic of the FEA Analysis ................................................... 29
Figure 14: Definition of Material Properties in Engineering Data .............................. 30
Figure 15: Rim Flange, Outer Spoke and Outer Hub of the 1098 Ducati Rim ........ 32
Figure 16: Inner Spoke and Inner Hub of the 1098 Ducati Rim ................................ 32
Figure 17: Rim Flange Cosmetic, Spoke to Spoke Re-enforcement, Outer Spoke to
Rim Re-enforcement of the 1098 Ducati Rim ........................................................... 33
Figure 18: Inner Spoke to Rim Re-Enforcement and Inner Spoke to Outer Spoke Re-
Enforcement ............................................................................................................. 33
Figure 19: Example of Co-ordinate Systems of Components................................... 34
Figure 20: Example of Co-ordinate System for Spoke to Spoke Re-enforcements .. 34
Figure 21: Material of 1098 Ducati Setup in ACP Pre Setup .................................... 35
Figure 22: ACP Pre Process: Modelling Ply Groups ................................................ 35
Figure 23: Sampling Element in the Hub Section ..................................................... 36
Figure 24: Illustration of Orientated Element Sets of Rim Flange ............................. 37
Figure 25: Illustration of Reference Fibre Direction (Yellow Arrow) and Fibre
Direction (Green Arrow) ........................................................................................... 37
Figure 26: Thickness Variation: Ducati Rim ............................................................. 38
Figure 27: Thickness Variation: Outer Section-Ducati Rim ...................................... 38
Figure 28: Meshed Imported Layered Section of 1098 Ducati Rim (Coarse Mesh) . 39

viii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Figure 29: Refinement of the Ducati Rim Mesh ........................................................ 40
Figure 30: Illustration of Fixed Support on the Ducati Rim ....................................... 41
Figure 31: Remote Loading of the Ducati Rim ......................................................... 41
Figure 32: Longitudinal Strains: Outer Section - Ducati Rim .................................... 43
Figure 33: Longitudinal Strains: Inner Section - Ducati Rim ..................................... 44
Figure 34: Transverse Strain: Outer Section - Ducati Rim ....................................... 45
Figure 35: Transverse Strains: Inner Section- Ducati Rim ...................................... 46
Figure 36: In Plane Shear Strain: Outer Section - Ducati Rim.................................. 46
Figure 37: In Plane Shear Strain: Inner Section - Ducati Rim .................................. 47
Figure 38 : Defining the Maximum Failure Criterion ................................................. 50
Figure 39: IRF - Longitudinal Strain Mode: Outer Section - Ducati Rim ................... 51
Figure 40: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim .................... 51
Figure 41: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim .................... 52
Figure 42: IRF - Transverse Strain Mode: Outer Section - Ducati Rim .................... 52
Figure 43: IRF - Transverse Strain Mode: Inner Section - Ducati Rim ..................... 53
Figure 44: IRF - Transverse Strain Mode: Inner Section - Ducati Rim ..................... 53
Figure 45: IRF- In Plane Shear Strain Mode: Outer Section - Ducati Rim ................ 54
Figure 46: Maximum Strain - In Plane Shear Strain Mode: Inner Section - Ducati Rim
................................................................................................................................. 54
Figure 47: IRF - In Plane Shear Strain Mode: Outer Section - Ducati Rim ............... 55
Figure 48: Load Case of Passenger Vehicle Carbon Fibre Rim ............................... 57
Figure 49: Longitudinal Strains: Outer Section - Ducati Rim (2000 Nm) .................. 58
Figure 50: Longitudinal Strains: Inner Section - Ducati Rim (2000 Nm) ................... 58
Figure 51: Transverse Strains: Outer Section - Ducati Rim (2000 Nm) .................... 59
Figure 52: Transverse Strains: Inner Section - Ducati Rim (2000 Nm)..................... 59
Figure 53: In Plane Shear Strain: Outer Section - Ducati Rim (2000 Nm) ................ 60
Figure 54: In Plane Shear Strain: Inner Section - Ducati Rim (2000 Nm)................. 60
Figure 55: IRF- Longitudinal Strain Mode: Outer Section - Ducati Rim (2000 Nm) .. 61
Figure 56: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim (2000 Nm) .. 62
Figure 57: IRF - Transverse Strain Mode: Outer Section - Ducati Rim (2000 Nm)... 62
Figure 58: IRF- Transverse Strain Mode: Inner Section - Ducati Rim (2000 Nm) .... 63
Figure 59: IRF - In Plane Shear Strain Mode: Outer Section - Ducati Rim (2000 Nm)
................................................................................................................................. 63

ix
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Figure 60: IRF - In Plane Shear Strain Mode: Inner Section - Ducati Rim (2000 Nm)
................................................................................................................................. 64
Figure 61: Passenger Car Carbon Fibre Rim Hub Segments .................................. 65
Figure 62: Thickness Variation: Passenger Car Rim ................................................ 70
Figure 63: Thickness Variation: Outer Section- Passenger Car Rim ........................ 70
Figure 64: Refinement of the Passenger Car Mesh ................................................. 71
Figure 65: Longitudinal Strain: Outer Section - Passenger Car Rim ........................ 73
Figure 66: Longitudinal Strain: Inner Section - Passenger Car Rim ......................... 73
Figure 67: Transverse Strain: Outer Section - Passenger Car Rim .......................... 74
Figure 68: Transverse Strain: Inner Section - Passenger Car Rim .......................... 74
Figure 69: In-Plane Shear Strain: Outer Section - Passenger Car Rim .................... 75
Figure 70: In-Plane Strain: Inner Section - Passenger Car Rim ............................... 75
Figure 71: Maximum Strain Failure Criterion - Longitudinal Strain Mode: Outer
Section - Passenger Rim.......................................................................................... 77
Figure 72: IRF - Longitudinal Strain Mode: Inner Section - Passenger Rim ............. 78
Figure 73: IRF - Longitudinal Strain Mode: Inner Section - Passenger Rim ............ 78
Figure 74: IRF - Transverse Strain Mode: Outer Section - Passenger Rim............. 79
Figure 75: IRF - Transverse Strain Mode: Inner Section - Passenger Rim .............. 79
Figure 76: IRF - Transverse Strain Mode: Inner Section - Passenger Rim .............. 80
Figure 77: IRF - In Plane Shear Strain Mode: Outer Section - Passenger Rim ........ 80
Figure 78: IRF - In Plane Shear Strain Mode: Inner Section - Passenger Rim......... 81
Figure 79: IRF - In Plane Shear Strain Mode: Inner Section - Passenger Rim........ 81
Figure 80: Example of Strain Variation ..................................................................... 84
Figure 81: Stress Illustration on Cube [edited 45] .................................................... 95
Figure 82: Longitudinal Uniform Strain Loaded on RVE [edited 45] ......................... 96
Figure 83: Illustration of In-Plane Shear [edited 45] ................................................. 98
Figure 84: Illustration of Intra-Lamina Shear [edited 45]........................................... 99
Figure 85: Residual Strength Curves of the Composites [edited-32] ...................... 100

x
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

List of Tables

Table 1: Voltage Results at Zero Load of the 1098 Ducati Rim [8] ........................... 25
Table 2: Average Compression and Tensile Strain .................................................. 26
Table 3: Properties of the Carbon Fibre Plies [34] ................................................... 31
Table 4: Evaluation of Mesh Sizing .......................................................................... 40
Table 5: Comparison of FEA and Experimental Strains ........................................... 48
Table 6: Example of Bending Moment Calculation for the Vehicle ........................... 56
Table 7: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Hub Section 66
Table 8: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Spoke Section
................................................................................................................................. 67
Table 9: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Rim Section 68
Table 10: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Rim Re-
enforcement ............................................................................................................. 69
Table 11: Evaluation of Mesh Sizing (Passenger Car Rim)...................................... 71
Table 12: Comparison of the FEA strain of Passenger Car and Ducati Rim ............ 76
Table 13: Comparison of the Ducati (2000 Nm) and Passenger Rim Strains........... 76
Table 14: Comparison of the Ducati (2000 Nm) and Passenger Rim IRF ................ 82
Table 15: Thickness Comparison of the Rims .......................................................... 86

xi
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

List of Symbols

A Total area (mm2 )

Am Matrix area (mm2 )

Af Fibre area (mm2 )

Ef Fibre modulus (MPa)

E1 Longitudinal modulus (MPa)

E2 Transverse in-plane modulus (MPa)

E3 Transverse out of plane modulus (MPa)

Em Matrix modulus (MPa)

𝐹 Force (N)

𝐹𝑟 Permissible static wheel load (N)

Gf Fibre shear modulus (MPa)

Gm Matrix shear modulus (MPa)

G13 Shear modulus through the thickness (MPa)

G12 In-plane shear modulus (MPa)

G23 Shear modulus in the plane perpendicular to the fibre (MPa)

𝑀 Moment (Nm)

𝑀𝑏𝑚𝑎𝑥 Maximum bending moment (Nm)

𝑁𝑓 Fatigue life

Vf Fibre volume percentage (%)

Vm Matrix volume percentage (%)

𝑑 Offset (m), the distance between the hub and the rim centreline

xii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
𝑚 Mass (kg)

𝑟𝑑𝑦𝑛 Dynamic tyre radius of largest intended tyre (m)

𝑟 Radius (mm)

𝜖𝑚 Matrix strain

𝜖𝑓 Fibre strain

𝜀1 Longitudinal strain

𝜖11 Strain in 11 direction

𝜀2 Strain in 2-direction

𝜀12 In plane strain

𝑓 Wheel load increase factor

σf Fibre stress (MPa)

σm Matrix stress (MPa)

𝜎1 Longitudinal stress (MPa)

𝜎12 In plane stress

𝜎2 Transverse stress (MPa)

𝜎1𝑡 Longitudinal tensile strength (MPa)

𝜎1𝑐 Longitudinal compression strength (MPa)

𝜎2𝑡 Transverse tensile strength (MPa)

𝜎2𝑐 Transverse compressive strength (MPa)

𝜏12 In plane shear (MPa)

𝑣12 In-plane Poisson`s ratio

𝑣𝑓 Fibre Poisson`s ratio

xiii
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
𝑣𝑚 Matrix Poisson`s ratio

𝜇 Coefficient of friction
𝛾12 Shear strain

𝜃 Strain gauge angle

xiv
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 1: Introduction

1.1 Background

The wheel (rim) is a critical component in its form and function with respect to vehicle
motion in the automotive industry [1]. The wheel is used in major forms of land
transport (passenger vehicles, heavy industrial vehicles and motorcycles) and no
one can dispute the importance of the wheel. The main function of the wheel is to
transfer energy from the vehicle to the tyre [2]. The wheel assists in vehicle support,
the comfort of the vehicle ride and torque transfer during braking and acceleration
[2].

In the last decade, wheel manufacturers have focused on lightweight wheel designs.
This was done by introducing new materials and manufacturing technologies, which
are in line with the changing requirements in the automotive industry [3]. According
to the above mentioned functions it is evident that the wheel is so important and
ubiquitous, that designing an improved wheel from a new and more advanced
material would have a major benefit on the transport industry. The major benefit for
the transport industry will amount to fuel saving due to the high increase in fuel
prices. The use of lighter materials with the required stiffness in vehicles will result in
decreased fuel usage [4]. The main focus point of this research is to investigate the
feasibility of a carbon fibre rim for a passenger vehicle.

Figure 1: 1098 Ducati BlackStone Tek Rims [5].

1
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
This research is a continuation of the investigation that evaluated the novel use of
carbon fibre in designing an automotive car rim by Sali et al. [6]. The investigation
looked at adapting the 1098 BST (BlackStone Tek) motorcycle rim (which is
illustrated in Figure 1).

Grech-Cumbo [7] investigated the heating of the carbon fibre rim from the brakes of
a car. A steel cylinder the same size as a brake disc was heated to 600 °C in a
furnace and placed onto the hub of the BST 1098 motorcycle rim to simulate the
heating of the rim from the brakes. The carbon fibre showed some discolouration
and further investigation suggested to understand and quantify the heating of the
carbon fibre from the car`s brake discs. Further work was required to ensure that the
structural strength of the carbon fibre is not compromised by the heat from the
brakes of the car.

Sali et al. [6] investigated the structural integrity of the rim required to pass the
dynamic cornering fatigue test requirements (the wheel testing concepts will be
discussed in further detail in section 1.3). The computational calculations showed
that the BST 1098 rim would be strong enough to support the increased loads of a
car by adding more layers of carbon to the spokes and by redesigning the aluminium
hub. The FEA (finite element analysis) calculations were not validated with
experimental results.

The focus of this research report is to investigate if carbon fibre rims are feasible for
use on passenger cars. In 2010, Kienhöfer [8] performed the cornering fatigue test
on the BlackStone Tek Ducati rim; to simulate the strains experienced during the
test. The strain results will be utilised to validate the FEA model of the BlackStone
Tek Ducati rim and investigate the feasibility of a carbon fibre passenger rim. The
Ducati rim will be further discussed in the section below.

2
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
1.2 BlackStone Tek Ducati Rim

Wheel design has seen an evolution in material, from polymeric composites (wood)
to metals (steel and aluminium), to carbon fibre as illustrated by BlackStone Tek
(BST). BST is a South African company that designs, manufactures and tests
composite motorcycle rims. BST is a ISO 9001:2008 certified company, and
manufactures aftermarket rims that can be used in competitive motorcycle racing [5].
The rims are manufactured as one unit using the autoclave manufacturing process.
This process assists in the manufacturing of the rim through regulation of the
pressure and temperature in the envelope of the autoclave oven, which is essentially
a pressure vessel [3]. Rim Technology, Eurotype Test Centre and German TÜV
have tested the wheels and they have met all requirements with regards to road
certification [5].

The 1098 Ducati rim (Figure 1) weighs 2.7 kg. The composite motorcycle rims
reduce the mass and rotating moment of inertia, which significantly reduces the
forces required to move or steer the motorcycle [5]. Therefore the lighter wheel
improves handling, acceleration and deceleration, which can be attributed to the
mass distribution in the wheel [5]. The composite rims also offer excellent aesthetic
appeal.

Figure 2 illustrates the components of the BST Ducati rim. These components assist
in the basic functionality of the rim (wheel) and the fitment of the tyre to the rim
section. The wheel loads are dependent on the vehicle characteristics (tyre pressure;
weight; suspension system etc.) and service conditions [9].

3
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 2: Anatomy of the BST Ducati Rim

The wheel`s main components are:

 Hub: the attachment point between the wheel and the vehicle. This section
consists of bolt holes along the bore diameter to facilitate in the fastening of
the wheel to the vehicle [2].
 Rim: the interface between the wheel and tyre, that assists in tyre fitment [2].
 Spokes: the structural members that transfer the loads from the hub to the
rim.

The 1098 Ducati BST rim as indicated in Figure 2, will be the reference design for
the development of the carbon fibre passenger vehicle rim. The BST motorcycle rim
is used as a benchmark design, due to the moulds being expensive (approximately
R250 000 each). The same motor cycle rim moulds will be used to manufacture the
carbon fibre passenger vehicle rims, once the structural integrity checks have been
done. If the integrity checks give a positive result; then the secondary problem of the
heating of the carbon fibre can be evaluated and possible solutions developed. The
wheel testing specification will be discussed in the section 1.3.

1.3 Wheel Test Methods

4
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The structural integrity of the wheel can be evaluated using the test methods outlined
in Figure 3. The tests evaluate the accelerated fatigue life of the wheel. Wheels that
are intended for passenger use must pass an impact test, cornering fatigue test and
radial fatigue test before they go into production [3].

Figure 3: Wheel Test Methods [10; 11]

 Radial Fatigue Test: A wheel with a tyre is loaded against a rotating drum to
simulate straight ahead driving conditions [10]. Accelerated loads are applied
to evaluate stresses into the tyre and rim [10].

 Cornering Fatigue Test: The cornering fatigue test entails constraining the
rim flange while a force is applied to the hub through a load arm [12]. The test
simulates cornering driving conditions.

 Impact Test: The rim flange of a wheel and tyre assembly is impact loaded
[11]. The assembly is mounted at a 13 degree angle to the horizontal plane,
before the load is applied. This test simulates the loading experienced when a
vehicle drives over a pavement or experiences a sudden impact.

The failure criterion for each of the tests is considered to be any permanent
deformation of the wheel or wheel assembly (tyre and wheel). This ranges from
crack initiation or loss of tyre pressure due to wheel deformation depending on the
test in question. The carbon fibre rim will be structurally validated through the
cornering fatigue test with reference to the Society of Automotive Engineers (SAE)

5
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Handbook J328 [13] as illustrated in Figure 3. Karandikar et al. [14], Raju et al. [3],
Guo et al. [15], Fischer et al. [16] , Kocabicak et al. [17] , Riesner et al. [18] and
Wang et al. [19] [12] have compiled research papers outlining the simulation of the
dynamic cornering fatigue tests of metal (steel and aluminium) automotive wheels.
The papers outlined above overlook the dynamic cornering fatigue test (CFT) of
carbon fibre rims, as they are intended for metallic rims, but are relevant with regards
to carbon fibre rims. The CFT has the highest failure rate amongst the three tests
[3]. Therefore the CFT is the relevant test with regards to the research investigation.

1.4 Problem Description

The automotive industry is a crucial industry of the economy. The retail sales of
automotive fuel in 2009 formed 29% of the income of R365 906 million of the motor
trade industry in South Africa [20]. Retail sales of motor vehicle and sales of new
motor vehicle parts and accessories are 49% and 9% respectively as indicated by
Statistics South Africa [20]. It is evident that wheels (rims) play a crucial role in
everyday life, due to the size of the motor trade industry, which contributes 6% to the
GDP of South Africa [14]. The industry is a significant size, but carbon fibre rims for
passenger cars have a non-existent share of the market. Steel and aluminium rims
currently dominate the market [14] and therefore the development of a carbon fibre
passenger car rim (wheel) will stimulate additional market growth in the motor trade
industry.

In the development of new material for wheels (rims), it is essential that fatigue
testing be conducted on the wheels in order to validate the wheel`s structural and
material integrity. The material shift from the conventional steel to aluminium wheels
in the late 90s was part and parcel of the research done to develop fatigue curve
results for aluminium wheels [10]. Carbon fibre is an extremely strong and light
material. Its use as a suitable material to manufacture passenger car rims will be
investigated in this report.

6
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
1.5 The Proposal

BlackStone Tek in partnership with the University of the Witwatersrand will be

evaluating the feasibility of using carbon fibre to manufacture a passenger car rim.
The 1098 Ducati rim will be used as a reference design for the carbon fibre
passenger car rim as discussed earlier. The 1098 Ducati rim CFT strain results
obtained by Kienhöfer [8] will be the starting point of the research. An FEA model of
the 1098 Ducati rim will be developed to simulate the experimental CFT conditions.
A good correlation between the Ducati rim`s dynamic cornering fatigue experimental
and FEA strain results will validate the finite element technique.

The finite element technique will be utilised to develop the passenger vehicle rim
FEA model. The structural integrity will be evaluated on strains on the wheel. The
wheel`s ability to pass the cornering fatigue test will determine the fatigue life and
durability of the rim [18].

In the next section/chapter the literature, apparatus and tests required to design and
structurally verify a wheel design will be covered. This will include experimental and
finite element analysis.

7
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 2: General Theory

Wheels have been in existence since the Sumerian civilization of Mesopotamia [21].
The wheel has evolved from wooden chariot wheels; to wire spoke wheels; and
finally steel and aluminium wheels [10]. Initially the fatigue failure of wheels was not
evaluated due the low speeds and short vehicle life spans [10]. Once the vehicles
could travel greater distances and reach faster speeds, then fatigue analysis of
wheels became a priority. Fatigue analysis requires an understanding of the form,
function and material of the wheel.

The foundation of design is in line with “the form follows function” principle as stated
by Louis Sullivan an American architect [22]. The form follows material principle is a
build on to the form follows function principle and addresses material design with
respect to products structural form [22].

Form follows function is influential when it comes to the technical aspects of rim
design. These aspects can be attributed to the rim components (Figure 2); rim
shape; mass distribution within the rim (wheel) sections; moment of inertia of rim and
the load cases experienced during braking; cornering and acceleration.

Figure 4: Wheel Dynamic Variables [23]

The dynamic load cases on the wheel are illustrated in Figure 4. The automobile
weight is shared between the front and rear wheels. The weight distribution and
suspension configuration on the vehicle will determine the load on each wheel. In

8
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
general the front wheels are used as the design reference point, in that they sustain
greater load compared to the rear wheels. The wheel will experience lateral and
longitudinal tyre force during cornering, braking, deceleration and acceleration
through the tyre. The rim (wheel) should be able to handle these loads and maintain
its structural integrity.

The material selection is in line with minimising the weight of the rim. The reduced
density and volume properties of the material will determine the mass of the
composite wheel.

The reduction in weight due to carbon fibre usage will improve the strength to weight
ratio and minimise the moment of inertia [5]. The aim is to ensure that the centre of
gravity coincides with the geometric centre of the wheel. Composite rim allows for
the reduction in the overall weight in comparison to aluminium rims [5]. Therefore
mass distribution and overall weight can reduce the moment of inertia and resistance
to angular acceleration.

Form follows material is influential in material design with respect to composite rims.
The general rule is that material should be used in the most efficient way, which
allows for their properties and material malleability to benefit the product form and
ultimately the function [22].

The structural integrity of the rim is dependent on the loading, manufacturing

process, the mechanical properties and allowable stresses of the material [8].
Classical micromechanics as outlined in appendix A will assist in defining the
properties of the composite material. The stiffness properties of the wheel are
dependent on the material micromechanics. This fundamentally determines the rim`s
ability to resist deformation. The specification of CFT is outlined in section 2.1.

2.1 Design Standards for Wheels

Design standards of the wheels analyse the wheel fatigue strength by simulating
load cases (radial and cornering loads) on a wheel during road usage [10]. The load
cases are cyclic and encompass different load spectrums considering the load
amplitude and frequency, which is dependent on the vehicle motion (braking,
cornering, decelerating and accelerating) [24]. SAE (Society of Automotive

9
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Engineers) J328 for passenger car and light truck steel wheels is the industry
standard for fatigue test specification [10]. The SAE test specification has since
evolved to take into account material changes and application changes in the wheels
as illustrated in specifications such as SAE J1204 (trailer applications), SAE
328/J267 (temporary compact spares usage wheel) and SAE J1992 (military
applications for take apart wheels) [10].

ISO 3006: 2005 (Road vehicles—Passenger car wheels for road use—Test
methods) and ISO 3894:2005 (Road vehicles—Wheels/rims for commercial
vehicles—Test methods) are based on SAE specifications [10]. The specification
assisted in the development of material fatigue curves for wheels based on the
material`s condition and the type of loading under study. The results were later
normalised to the ultimate fatigue life required from the wheel in service [10].

Figure 5: 1098 Ducati Rim Setup for Cornering Fatigue Test

The loading conditions that the rim will be exposed to during the cornering fatigue
test are highlighted in Figure 5. The rim flange is constrained while a force is applied
to the hub at a moment’s arm through a rigid shaft, during the rotation of the rim [10].
During the test the rim is fitted with strain gauges in order to evaluate and quantify
the strains. This is very important in product verification. In accordance with SAE
specification the constant force needs to be applied at a minimum distance of 30 in

10
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
(762 mm) [10]. The deflection is measured at the loaded pointed on the shaft and the
load is parallel to the plane through the centre of the rim as illustrated in Figure 5
[10]. In the detail rim design the dynamic cornering fatigue test is critical. The fatigue
properties of the wheel (rim) is the ability of the rim to be subjected to a bending
moment for 100000 cycles without displaying any evidence of harmful cracks,
significant deformation or any abnormal looseness at joints [25]. This is applicable to
specifications such as TÜV, ISO 3006:2005 and SAE Handbook J328.

Load formula

The maximum bending moment experienced by the rim during cornering fatigue test
is expressed in equation 2.1 [10].

𝑴𝒃𝒎𝒂𝒙 = 𝒇𝑭𝒓 (𝝁𝒓𝒅𝒚𝒏 + 𝒅)................................................................................ (Eq. 2.1)

Where:

𝑀𝑏𝑚𝑎𝑥 Maximum bending moment [Nm]

𝑓 Wheel load increase factor is 1.5 which can be obtained by

standards [17].

𝜇 Coefficient of friction between the tyre and the road.

Drop centre rims (μ = 0, 7) and bead seat rims (μ = 0, 65) [26].
𝑟𝑑𝑦𝑛 Dynamic tyre radius of largest intended tyre [m]

𝑑 offset [m]. The distance between the hub and the rim

centreline as illustrated in Figure 2.

𝐹𝑟 Permissible static wheel load [N]

The assumption is that the passenger car total weight, consists of the passengers,
vehicle and luggage load. The front wheels each support 30% of the total load and
the rear wheels support 20% each. Therefore the front wheels are analysed as they
experience the largest load. This is considered to be the permissible static wheel
load. The CFT will be utilised to test the rim experimentally and through FEA. The

11
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
experimental analysis was done using strain gauges to obtain the longitudinal
directional strains due to their versatility and precision of measurement [2]. The
strain gauge results will be converted into longitudinal, transverse and in plane shear
strain using the formulas in section 2.2.

2.2 Experimental Analysis of the Ducati Rim

The validity of the finite element technique will be dependent on the experimental
strain results. The experimental testing will lay the foundation for all the FEA
modelling to follow. The section 2.2 illustrates an overview of the stress-strain
relationship and the principles of strain gauging. Classical micromechanics were the
fundamental basis of this report as illustrated in appendix A.

2.2.1 Stress strain relationship of a lamina

A lamina is a single layer of carbon fibre. The lamina is assumed to be stress free,
void free and perfectly bonded [28]. A laminate is a combination of two or more
bonded laminate [27]. The strain stress relationship is based on the assumption
listed below:

 The matrix, fibre and lamina is isotropic and linearly elastic [27]

Figure 6: Unidirectional Fibres [28]

The strain stress relationship in the principal directions is illustrated in the matrix
formula in equation 2.2 [27]. Hooke`s law stress-strain relationship in a 1-2-3

12
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
orthogonal system forms a six by six matrix indicating 6 stresses, 6 strains and 6
displacements. Equation 2.2 is a simplified 3 by 3 matrix, because a lamina is a thin
layer and can be modelled using a plane stress model. The two dimensional analysis
or plane stresses will assist in analysing surface stresses and strains in two
dimensions [2].

1 𝑣12
𝜀1 − 0 𝜎1
𝐸1 𝐸1
{ 𝜀2 } = −𝑣21 1
0 { 𝜎2 } ................................................................... (Eq. 2.2)
𝛾12 𝐸2 𝐸2 𝜎12
{ 0 0 𝐺12 }

The stiffness matrixes are symmetrically as illustrated in equation 2.2 [28].

𝑣12 𝑣21
= ......................................................................................................... (Eq.2.3)
𝐸1 𝐸2

Where:

E1 Longitudinal modulus (MPa)

𝜀1 Longitudinal strain

𝜎1 Longitudinal stress (MPa)

E2 Transverse modulus (MPa)

𝜎2 Transverse stress (MPa)

𝑣12 In-plane Poisson`s ratio

G12 In-plane shear modulus (MPa)

𝛾12 Shear strain

𝜎12 In plane stress (MPa)

The two dimensional analysis will assist in linking up to the strain data obtained in
the experimental testing. The principles of strain gauging will be discussed in section
2.2.2.

13
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
2.2.2 Principles of strain gauging

Strains are related to a small change in electrical resistance within a Wheatstone

bridge circuit as indicated in Figure 7 [2]. The individual strain gauges measure
longitudinal strains. The individual strains are transformed from the 1-2 coordinates
into the 11 coordinate using equations 2.4 [29]. The engineering shear strain (𝛾12) is
twice the tensor shear strain (𝜀12 ).

Figure 7: Strain Gauge Rosette [29] and Wheatstone bridge Circuit [2]

𝜀1 +𝜀2 𝜀1 −𝜀2
𝜀1′ = + cos 𝜃 + 𝜀12 sin 𝜃............................................................... (Eq. 2.4)
2 2

Where:

𝜀1′ Longitudinal strain (11 – direction)

𝜀1 Strain in longitudinal-direction (1 – direction)

𝜀2 Strain in transverse-direction (2 – direction)

𝜀12 In plane shear strain

𝜃 The angle between axis 1 and axis 11

14
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The wheatstone bridge circuit and strain gauge were used to evaluate the surface
strains of the Ducati rim. The experimental strain will be compared to the FEA strain.
The literature review is highlighted in chapter 3, to evaluate relevant research in line
with the subject matter.

15
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 3: Literature Review

This section outlines the experimental analysis and numerical analysis of rims with
respect to the cornering fatigue test. The detailed rim design entails understanding
the loading conditions, material design and structural design. The design was
validated through experimental and numerical methods.

3.1 Experimental Analysis

The SAE (Society of Automotive Engineers) J328 for passenger car and light truck
steel wheels is the industry standard for fatigue test specification [10]. An excellent
and a concise overview of the development of experimental fatigue tests for wheels
was compiled by Kinstler [10]. One test method is the radial loading of a wheel
assembly (tyre on the wheel), which is at a camber angle relative to the rotating
drum [10]. This method is a build on to the straight driving radial testing, but in this
case it was meant to simulate cornering driving conditions. The wheel assembly is
loaded against the rotating drum using camber angles ranging from 5 o to 30o [10].
This test proved to be difficult to execute; due to the low tyre life, which required
frequent tyre changes to complete the test [10]. The test made it difficult to evaluate
the wheel for cracks and deflection [10].

A second approach is to clamp the wheel assembly to a rotating table and apply a
bending moment to the wheel (rim), through a shaft attached to the hub. Increased
shaft deflection, fatigue cracking on the wheel and a decreased clamping force
(loose clamps and nuts) is used as indication to terminate the test [10]. The method
allows for easy and fast testing execution. The absence of the tyre allowed for higher
test factors and cycles of wheel rotation, but these settings varied from each wheel
manufacturer [10]. Therefore the SAE standardised the test equipment and
procedure such as loading and cycles.

Muthuraj et al. [1] carried out experimental stress analysis on aluminium alloy wheels
(original and modified design) using strain gauges to capture strains during the
dynamic cornering fatigue test. Stresses were calculated through Hooke`s law of

16
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
elasticity using the strains obtained through the strain gauges [1]. The elastic strain
and isotropic Young`s modulus in the wheel allows for the use of Hooke`s law.

Figure 8: 45 Degree Rosettes Instrumented on the Wheel [edited 11]

Wang et al. [12] simulated the dynamic cornering fatigue test of a steel passenger
car wheel [12]. The strain data are obtained from the instrumented wheel at points A,
B, C, D and E as illustrated in Figure 8. The points are instrumented with 45 degree
rosettes, which consist of three strain gauges orientated at 0, 45 and 90 degrees
[12]. The strains on the wheel were measured at four positions (0, 90, 180 and 270
degree orientations) during the cornering fatigue simulation, where the force
application was aligned with these positions as illustrated in Figure 8 [12].

The captured strains at the points were utilised to calculate the von Mises stresses
[12]. The strain data illustrated the strain variation with the position of the wheel and
the application of the force [12]. The data indicate a sinusoidal wave relationship
between the strains obtained and the position of the wheel and force application.
This relationship was expected as the largest strains are experienced only when the
orientation of the strain gauges was in line with the force application path. The local

17
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
strain approach was used to predict fatigue failure of the wheel, using the stress
strain curve. The experimental fatigue testing of metallic wheels is illustrated in this
section, but relevance can be drawn with regards to carbon fibre rims. The Ducati
rim was subjected to CFT conditions and instrumented with strain gauges to
evaluate the longitudinal strains. The highlighted research papers evaluate the
experimental data of metallic rims as intended; but research papers evaluating
carbon fibre rims for passenger vehicles are limited. The numerical (FEA) analyses
of wheels will be discussed in section 3.2.

18
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
3.2 Numerical Analysis

Karandikar et al. [14] developed FLIPS, which is a prototype computer based system
that can predict the fatigue life of wheels. The FEA package is able to simulate the
rotating bending moment fatigue test to obtain the lower bound on the fatigue life and
the critical areas of the wheel, which is based on the local strain approach [14].

Guo et al. [15] outlined the FEA procedure to evaluate the fatigue life of a cast
aluminium wheel using the S-N approach for a constant reverse loading condition
[15]. The wheel was fatigue tested with reference to SAE J328, to evaluate the effect
the clamp loads inclusion and exclusion have with regards to fatigue life predictions.
The research illustrated that to accurately evaluate fatigue life prediction close to bolt
locations, it was essential to include the mean stress changes induced by the
clamping force [15].

Fischer et al. [16] evaluated the critical parameters to design a durable light-weight
wheel (hub) design. The research indicated that the operational loading (straight
driving and cornering stress) required to test the fatigue properties of the hub, which
are dependent on the material and manufacturing process.

Hsu et al. [30] described the prediction of fatigue failures of aluminium disc wheels
using the failure probability contour based on historical test data [30]. The historical
data of dynamic cornering fatigue test were linked with simulation results to construct
probability failure contours [30]. The finite element simulations captured the mean
stress and stress amplitudes that were plotted on a two-dimensional plane [30].
Therefore for a new wheel, the failure probability of dynamic cornering fatigue test
can be read directly from this probability contour [30].

Kocabicak et al. [17] performed a wheel numerical analysis using the bi-axial load-
notch strain approximation for proportional loading [17]. The method estimated the
fatigue life of a passenger car wheel during the cornering fatigue test under plane
stress conditions [17]. The total deformation theory of plasticity evaluated the elasto-
plastic strain of the component and the Palgren-Miner rule was the applicable
damage accumulation method [17].

19
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Wang et al. [19] proposed a fatigue failure life prediction method of aluminium
wheels that integrates finite element analysis and the nominal stress method [18]. A
static load finite element model of the aluminium wheel simulated the rotary fatigue
test to obtain the equivalent stresses [19]. The equivalent stress was calculated
using the nominal stress method based on effects such as the mean load size,
fatigue notch, surface finish and scatter factors [19]. The fatigue life of the aluminium
wheel was predicted using the equivalent stress amplitude and aluminium alloy
wheel S-N curve [19]. This method resulted in accurate correlation between the
bench test and the finite element model.

Fleischhauer et al. [31] evaluated the non-linear simulation of a lightweight

composite rim. The relevance will be drawn from the finite element modelling
performed on the Formula Student Race vehicle wheel. The initial design entailed
the optimisation of the isotropic rim geometry, through evaluation of main stresses
and local stress maxima induced by the maximum side and vertical load [31]. The
optimised geometry allowed for evaluation of main stress plots, which allowed for
optimisation of the fibre orientation and laminate structure [31]. A FEA model with
layered shell elements was developed from the optimised geometry. The layered FE
model is parameterised according to constant component thickness and constant
layer structures [30]. The prepregs used in the build-up of the rim are unidirectional
HT carbon fibre and woven fabric prepreg. The high precision Schürmann/Puck
criterion was used in the analysis of fibre failure and inter-fibre failure [31]. The
criterion indicates layers exposed to high magnitude loads, which could lead to fibre
failures [31]. The highly stressed region is the transition between the spoke and rim
bed and hub [31]. The numerical fatigue testing of metal and composite wheels is
illustrated in this section, but research papers evaluating carbon fibre rims for
passenger vehicles are limited. Previous work on composites is highlighted in
section 3.3.

20
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
3.3 Previous Work on Composites

Composite material design assists in determining the limits of applicability of the

material itself. This is work done with respect to the design, development and
manufacture of composites [32]. Rim technology has evolved with material, due to
the need for lighter rims. Giek et al. [33] examined and quantified fuel economy and
energy incentives for the use of light-weight wheels on passenger cars [30]. The
article evaluates the use of glass reinforcement plastic wheels and a laminated,
form-filled wheel. Similar work has been done in improving the fundamental
properties of line-based carbon fibre for transport applications by Compere et al. [4].
The research was done to evaluate the use of carbon fibre in automobiles in order to
decrease fuel consumption and the overall weight of the vehicle.

Composites are being explored in the ultimate aim to replace steel and aluminium
rims. Fatigue life prediction, experimental and FEA methods have assisted in the
advancement of material selection with regards to rims. Fleischhauer et al. [31]
evaluated the non-linear simulation of a lightweight composite rim. This paper
evaluates a carbon fibre replacement rim for aluminium rim for a Formula Student
racing car. The carbon fibre rim is validated through finite element analysis.
BlackStone Tek has also done extensive development into composite motorcycle
rims. The objectives of the research are highlighted in chapter 4.

21
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 4: Objectives

4.1 Research Scope

The research report investigates the feasibility of a passenger car carbon fibre rim.
The feasibility is to be validated through experimental and numerical cornering
fatigue test data. In addition, the passenger car carbon fibre rim will be compared to
its aluminium counterpart to quantify the weight saving.

Since the moulds required to manufacture the rim are expensive, the passenger car
rim will be adapted to the 1098 Ducati rim mould for manufacturing purposes. The
finite element model of the composite passenger car rim will be based on the
correlation between the experimental and the FEA strain results of the 1098 Ducati
BST rim, during the cornering fatigue test . The correlation accuracy between the two
evaluation methods will be used to optimise the finite element analysis (FEA) of the
carbon fibre passenger car. The investigation steps are outlined below:

1. Calculate the numerical strains, through the FEA modelling on the 1098
Ducati rim. This will be done by simulating the dynamic cornering fatigue test
conditions experimentally tested by Kienhöfer [8] in order to validate the FEA
modelling.
2. Investigate the correlation between the numerical and experimental results, to
validate the FEA method. An accurate correlation will affirm the accuracy of
assumptions, material inputs of the FEA modelling.
3. Design a carbon fibre rim for a passenger car using the 1098 Ducati rim as
the reference design and the validated FEA model.
The research outline is illustrated in section 4.2.

22
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
4.2 Research Outline

Chapter 5: In 2010, Kienhöfer [8] performed the cornering fatigue test on the
BlackStone Tek Ducati rim; to simulate the strains experienced during the test. This
chapter will review the experimental results. Equipment such as National Instruments
SCXI-1000 chassis, SCXI-1600 control card, SCXI-1520 universal card, SCXI-1314
module and strain gauges were used [8].

Chapter 6: The 1098 Ducati rim was evaluated using ANSYS. The model was
created using accurate carbon fibre layups of the rim. The strain results of the FEA
were correlated to the strain gauge results obtained experimentally.

Chapter 7: Design a carbon fibre rim to fit an automobile rim using the same 1098
Ducati mould and the validated rim FEA results from chapter six. The rim was
adjusted to handle higher loads and moments that would have been experienced in
the automobile. The rim design was validated through ANSYS.

Chapter 8: This chapter outlines the discussion, conclusions and the

recommendations.

23
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 5: Experimental Analysis: 1098 Ducati Rim

The experiment was conducted in 2010 by Kienhöfer at BlackStone Tek. The

experiment results are referenced from the strain gauge results of the 1098 Ducati
rim compiled by Kienhöfer [8]. The strain gauging by Kienhöfer was performed using
four channels, single 120 Ohm gauge in a quarter-bridge as illustrated in Figure 9.
The gauges were placed 80 mm from the inner circumference of the hub on the
respective spokes. The gauges were placed on the front face and the back face of
the respective spokes. One gauge was placed at 0 degrees (aligned to fiber direction
along spokes) and the other at 45 degrees (45 degrees to fiber direction along
spokes) [8]. The gauges were connected to an amplifier for data acquisition.

Figure 9: Strain Gauged Ducati Rim and Experimental Setup [edited 8]

The Ducati rim was subjected to CFT conditions highlighted in section 2.1. The
German TÜV test specification requires a bending moment of 540 Nm to be applied
[5]. The rim was subjected to a bending moment of 700 Nm, which is higher than the
specification. Six static zero load readings were taken for each strain channel in
Kienhöfer’s experiment, three before testing and three after to verify that the reading
did not drift. The readings were taken at a sample rate of 4000 Hz using a low pass

24
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
filter of 1000 Hz. The results of Kienhöfer’s investigation illustrate a sinusoidal
relation between the strain and the time (s) in Figure 10. Table 1 illustrate the
voltage results at zero loads during testing. The gauges were extremely stable and
the zero did not wander significantly [8].

Table 1: Voltage Results at Zero Load of the 1098 Ducati Rim [8]

Channel 0 Channel 1 Channel 2 Channel 3

Front 0 Front 45 Back 0 Back 45
0.0012 V 8.8578e-004 V 7.4628e-004 V 0.0012 V
Static 1
0.0012 V 8.8634e-004 V 7.6688e-004 V 0.0013 V
Static 2
0.0012 V 8.6590e-004 V 7.4182e-004 V 0.0013 V
Static 3
0.0012 V 8.7271e-004 V 7.4017e-004 V 0.0012 V
Static 4
0.0012 V 8.7280e-004 V 7.3760e-004 V 0.0013 V
Static 5
0.0012 V 8.7154e-004 V 7.2639e-004 V 0.0012 V
Static 6

Figure 10: Strain Gauge Results of 1098 Ducati Rim during One Test Run [8]

25
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Figure 10 illustrates that the strain gauges experience two peaks per revolution. The
peaks illustrate compression and tensile phases during the rotation of the rim. The
average tensile and compression strains are illustrated in Table 2.

Table 2: Average Compression and Tensile Strain

Channel 0 Channel 1 Channel 2 Channel 3

Strains Front 0 (µε) Front 45 (µε) Back 0 (µε) Back 45 (µε)
Average 173.8 254.2 120.4 448.9
Tensile
Average -165.2 -245.8 -137.6 -465.1
Compression

A sinusoidal relationship was visible between the strain and time as the wheel was
rotated. These finding were similar to the findings of Wang et al. [12], with respect to
the strain, time and the rotation of the wheel. In an ideal situation, the longitudinal
directional strains would be utilised to calculate the longitudinal, transverse, in plane
shear strains equivalent strains. The absence of a third longitudinal strain on the
front and back surface does limit the strain calculation (longitudinal, transverse and
the in plane shear strains). Therefore the longitudinal directional strains were
compared to the FEA longitudinal directional strains. The longitudinal, transverse
and the in plane shear strains obtained in the FEA were utilised to calculate the
longitudinal directional strains using equation 2.4. The FEA model of the Ducati rim
is evaluated in chapter 6.

26
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 6: FEA: 1098 Ducati Rim

This section aligns with the research report scope to investigate the numerical
strains of the Ducati rim. The FEA model of the current 1098 Ducati BST rim
simulated the cornering fatigue test. Please note that the experimental testing was
done with the aluminium hub attached to the rim, but the FEA analysis will only be
done with the rim. The detailed methodology followed during the FEA modelling is
outlined in section 6.1. Section 6.2 outlines the results of the current 1098 Ducati rim.
The sections are discussed in further detail below.

6.1 Methodology

The FEA methodology is outlined below, which entails model preparation, material
design and the loading for the rim. These aspects are discussed in further detail
below.

6.1.1 Model preparation

Figure 11: Surface Model of the 1098 Ducati Rim

The existing 1098 Ducati CAD model was imported into Solid Edge. The faces were
deleted on the model in order to convert it into a surface model as illustrated in
Figure 11. The conversion of the rim into a surface model allows for the definition of

27
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
the rim thickness, through material design in the FEA software platform. The
material design was done by setting the material properties, stacking sequence and
orientation of the plies. The rim was separated into three main components, which is
the hub section, spokes and rim section. The transition between these components
is crucial. It must be noted that the rim is manufactured in dies that are ultimately
joined during the autoclave process as illustrated in Figure 12. The rim consists of an
inner and outer section.

The transitions between the components are illustrated in detail A, B and C in Figure
11. These are the areas that require re-enforcement and the details are discussed
below:

- Detail A illustrates the transition from the spokes to the rim flange (bed). Re-
enforcement was required in this region to ensure proper dissipation of the
forces into the rim section from the spokes.
- Detail B illustrates the transition from the hub to the spokes. Since the hub
was highly stressed, it was essential to have re-enforcement in this area to
ensure no fatigue failures occur. Stress concentrations around the attachment
section (bolt holes) are expected.
- Detail C illustrates the transition from the inner spokes and outer spokes.
Reinforcement in this area was essential in joining both sections of the rim;
thereby ensuring the rim structural integrity.

Figure 12: Illustration of 1098 Ducati Rim Inner and Outer Sections

28
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Therefore the understanding of this geometry and stress plots was essential before
moving into the FEA (ANSYS) phase. Details A to C are illustrated in more detail in
section [Link]. This section illustrates the segmentation of the Ducati rim. The
surface model was an input into the ANSYS geometry section as illustrated in Figure
13.

6.1.2 Project schematic

The project schematic of the FEA analysis is illustrated in Figure 13. The process
flow is outlined in the project schematic. Static Structural is interfaced with ACP
(Ansys Composite Pre-Post) in order to clearly define composite properties. The
functional blocks of the project schematic are outlined below:

- Geometry: The parasolid of the rim surface model was exported from Solid
Edge as an input into this section. This section allows the user to define and
evaluate if the CAD model was within the required FEA specification.
- ACP (pre): This interface was utilised in the material design when setting
material properties, stacking sequence and orientation of the plies.

Figure 13: Project Schematic of the FEA Analysis

- Static structural: This interface was used in defining load cases, constraints
and the solution generation. This was essential in setting static conditions of
the cornering fatigue test.
- ACP (post): This Interface was utilised to evaluate the results, which in this
case will be the evaluation of strains experienced by the 1098 Ducati Rim.

29
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
6.1.3 Material design

Material design was important to the structural integrity of the rim. The essential
three factors in material (laminate) design of composites are:

- Material Properties: The macro/micromechanics and thickness of the material,

which are defined in the engineering data.
- Segmentation: The rim was segmented into components that make up the
rim. Material flows from segment to segment to form the entire rim as
illustrated in section [Link]. These components are unique and have their
own stacking sequence and ply orientation. Each component has a co-
ordinate system, which was used in the stacking sequences.
- Stacking Sequence: The orientation and placement of the plies was
dependent on the components co-ordinate system.

[Link] Material properties

Figure 14: Definition of Material Properties in Engineering Data

The material properties are defined in ACP (pre) engineering data as illustrated in
Figure 14. The materials that are defined in engineering data are T300 epoxy-

30
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
carbon, woven, 230 GPa Prepreg; T700 epoxy-carbon, woven, 230 GPa Prepreg;
and T700 epoxy-carbon, unidirectional, 230 GPa Prepreg. T300 was used for
aesthetics, but also plays a structural role in conjunction with T700 woven and
unidirectional. The properties listed in Figure 14 were extracted from Hexcel Prepreg
Technology and Torayca catalogue [34].

The properties of the respective materials are illustrated in Table 3. Unidirectional

materials are stronger in the longitudinal direction and weaker in the transverse
direction. Woven material is weaker than unidirectional material in the longitudinal
directional. The material is weakened by the fibre twisting (bending) during the
weaving process, but the woven material is stronger in the transverse direction in
comparison to unidirectional material.

Table 3: Properties of the Carbon Fibre Plies [34]

T300 Epoxy- T700 Epoxy-

Carbon, Carbon, T700 Epoxy-
Woven, 230 Unidirectional, 230 Carbon, Woven,
GPa Prepreg GPa Prepreg 230 GPa Prepreg
Density (kg/m^3) 1.8 1.8 1.8
E1 (GPa) 70 170 90
E2 (GPa) 70 10 90
G12 (GPa) 10 5 10
v12 0.1 0.3 0.1
Thickness 0.2 0.5 0.2

[Link] Segmentation

Segmentation of the rim was essential in the build-up of the 1098 Ducati rim. The rim
was segmented using named selections. This allows for geometric definition of the
parts on the rim. Part definition gives the user the freedom to define the co-ordinate
system for each part as illustrated in Figure 19 and Figure 20. The main components
of the rim are illustrated in Figure 15 and Figure 16. The main components are re-
enforced with re-enforcement components illustrated in Figure 17 and Figure 18. The
re-enforcement components assist in ensuring structural integrity in the transition
zones as discussed in section 6.1.1.

31
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 15: Rim Flange, Outer Spoke and Outer Hub of the 1098 Ducati Rim

Figure 16: Inner Spoke and Inner Hub of the 1098 Ducati Rim

The main components are the inner spoke, inner hub, outer hub, outer spoke and
rim flange section. The re-enforcement components are the rim flange cosmetic,
spoke to spoke re-enforcement, inner spoke to outer spoke re-enforcement
(intensifiers), outer spoke to rim re-enforcement and inner spoke to rim re-
enforcement. The stacking sequence of all the above mentioned will be defined in
section [Link]. The stacking sequence defined the properties of the main
components and transition (re-enforcement) zones.

32
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 17: Rim Flange Cosmetic, Spoke to Spoke Re-enforcement, Outer Spoke to
Rim Re-enforcement of the 1098 Ducati Rim

Figure 18: Inner Spoke to Rim Re-Enforcement and Inner Spoke to Outer Spoke Re-
Enforcement

Each part has a co-ordinate system, which assists in setting the reference fibre
direction as illustrated in Figure 24 . Co-ordinate system 01 was utilised for the rim
flange and rim flange cosmetics. The co-ordinate system was a cylindrical co-
ordinate system. Co-ordinate system 02 was utilised for the inner spokes, outer
spokes, inner spoke to rim-enforcement, outer spoke to rim enforcement and inner
spoke to outer spoke re-enforcement. The x–direction was along the length of the
spoke to ensure the fibres, align with load paths. A similar co-ordinate system was
duplicated along each spoke to match the orientation of each spoke.

33
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 19: Example of Co-ordinate Systems of Components

Figure 20: Example of Co-ordinate System for Spoke to Spoke Re-enforcements

Coordinate system 03 was used for the inner and outer hub. Coordinate system 04
was used for the spoke to spoke reinforcement. A similar co-ordinate system was
duplicated along each re-enforcement to match the orientation of each. Therefore it
was essential to have a co-ordinate system for each component (element set) as
discussed. The co-ordinate system was defined using rosettes in the ACP setup
which will be discussed in more detail in section [Link]. The components are
formed through named selections, which essentially lead to the formation of element
sets in the ACP setup. The stacking sequence of the components is outlined in
section [Link].

[Link] Stacking sequence

The stacking sequence was defined in the ACP Pre setup and this will be clarified in
this section. The actual layup of the rim will not be dealt with so as to protect the
proprietary knowledge of BlackStone Tek. The stacking sequence is illustrated in
Figure 22 in the modelling ply groups. The stacking sequence of the main
34
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
components and re-enforcement are illustrated in Figure 22 in the order they are
inserted into the layup. This is illustrated under the modelling ply groups. The
orientation and number of plies will not be dealt with.

Figure 21: Material of 1098 Ducati Setup in ACP Pre Setup

The material properties are extracted from engineering data, which lead to the
formation of fabrics. The fabric allows the user to enter the material thickness and
they can be used in the formation of stack-ups and sub laminates as illustrated in
Figure 21.

Figure 22: ACP Pre Process: Modelling Ply Groups

35
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The named selections are illustrated as elements sets in ACP. This assists in
defining the components as discussed in section [Link]. The co-ordinate systems
for the components are defined using rosettes as illustrated in Figure 22 . The
combination of the co-ordinate system and the element sets result in the formation of
orientated element sets as illustrated in Figure 24. The yellow vectors indicate the
reference fibre direction in the element sets. Orientated element sets were created
for all the components using the relevant co-ordinate system in section [Link]. It
was essential to ensure that the reference fibre directions for all the elements sets
are correctly defined, for accurate structural characterisation.

Figure 23: Sampling Element in the Hub Section

As an example, a sampling element was taken in the hub region as illustrated in

Figure 23. The sampling element was utilised to view the stacking sequence in the
hub region. The modelling ply is formed through a combination of material orientated
in different directions. The a (Figure 23) illustrates the angle of orientation of the ply
and the t (Figure 23) illustrates the ply thickness as illustrated in Figure 23.

36
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 24: Illustration of Orientated Element Sets of Rim Flange

The orientated element sets stacking sequence was modelled using modelling ply
groups, which allows the user to define the orientation and the quantity of stack-ups,
sub laminates and fabrics as illustrated in Figure 25.

Figure 25: Illustration of Reference Fibre Direction (Yellow Arrow) and Fibre
Direction (Green Arrow)

The ACP-Pre setup allows the user to define the stacking sequence in relation to
BlackStone Tek 9004 Layup. The completed defined product was imported into the
static structural section data. The stacking sequence gave the rim thickness and
geometric characteristics as illustrated in Figure 26 and Figure 27 . The rim varies
from a 4.1 mm hub, to 2.5 mm spokes and a rim section of 1.8 mm thickness
(beads excluded). The imported layered section of the 1098 Ducati rim is illustrated
in Figure 28.

37
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 26: Thickness Variation: Ducati Rim

Figure 27: Thickness Variation: Outer Section-Ducati Rim

38
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
6.1.4 Meshing

Figure 28: Meshed Imported Layered Section of 1098 Ducati Rim (Coarse Mesh)

Shell 181 elements were utilised in analysing the layers of the composite Ducati rim.
The shell elements are suitable for analysing thin to moderately thick shell structures
[35]. The element has 4 nodes with three transitional and three rotational degrees of
freedom [35]. The rim was meshed using the uniform quad element form. The use of
this element in its triangular form is not recommended especially in high stress
gradient areas [35]. The assumptions and restrictions of the shell 181 element in
accordance to Ansys are outlined below:

 Zero thickness elements are prohibited [35].

 No slippage is assumed between the element layers [35]
 Zero through thickness stress [35]
 Shell 181 elements are well suited for linear, creep, elastic and hyper-elastic
material properties [35].

39
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 29: Refinement of the Ducati Rim Mesh

The refinement of the mesh was crucial with respect to the accuracy of the results.
The spoke in line with the force application was the interest point as illustrated in
Figure 31. This spoke should experience the highest stress and strains. The mesh
on the spoke as illustrated in Figure 29 was refined to a finer mesh in comparison to
the rest of the rim geometry. The mesh sizing relevance centre and smoothing was
set to medium and the minimum element size was 0.5795 mm. The mesh used
69091 nodes and 69069 elements. The refinement of the illustrated sections in
Figure 29 is such that the mesh is two times smaller than the rest of the rim
geometry mesh. The refinement of the mesh caused a small change in the results as
illustrated in Table 4.

Table 4: Evaluation of Mesh Sizing

Medium Size
Strain Mesh Refined Mesh Difference
Longitudinal- 1Axis (µε) 929.18 929.57 0.04%
Transverse- 2 Axis (µε) 567.33 567.64 0.05%
In Plane Shear -12 Axis
(µε) -471.39 -471.61 0.05%

The discretization error and element size analysis was conducted on the refined
spoke region as illustrated in Figure 29. A strain probe was placed in the vicinity to
evaluate convergence of the results. The mesh was initially of medium size and later
refined, in order to compare the two stages. The refinement of the mesh caused a
small change in the results, thereby indicating that the sizing of the mesh is
appropriate. This illustrates that the results have converged and further mesh

40
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
refinements would only prolong the analysis without any benefit. The load cases are
illustrated in section 6.1.5.

6.1.5 Load cases

Figure 30: Illustration of Fixed Support on the Ducati Rim

The aim was to simulate the conditions the rim was exposed to during the
experimental testing. This allowed for the comparison of the dynamic cornering
fatigue experimental data with the static cornering fatigue FEA data. The static
conditions of the cornering fatigue test are illustrated in Figure 30 and Figure 31.
The rim was constrained on the rim flange as illustrated in Figure 30. The
constrained area is highlighted in blue in Figure 30. A 700 N remote force was
applied 1 m away from hub as illustrated in Figure 31. The force was attached to the
bolt holes on the inner and outer hub.

Figure 31: Remote Loading of the Ducati Rim

41
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The load conditions assisted in obtaining the results and the FEA analysis of the
Ducati rim are illustrated in section 6.2.

42
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
6.2 Result Evaluation: Current 1098 Ducati Rim

In this section the FEA results of the current Ducati rim are highlighted. The
longitudinal, transverse and in-plane shear strain are highlighted. The maximum
strain failure criterion was also evaluated. The main focus was on the strains
experienced on the spokes. The general strain pattern was also evaluated. The
results are presented below.

6.2.1 Strains

The strains from the hub are transmitted to the rim flange through the spokes. The
strains on the curvature of the spokes illustrate that the spokes are load bearing
sections. The strains around the spoke to rim flange transition illustrate that there is
a load transition from the spoke to the rim. The strain flow patterns assisted in
understanding the rim geometry and design. The following section evaluates the
longitudinal, transverse and the in-plane strains experienced during the static
analysis of the Ducati rim. The focus point was the spoke in line with the force
application as illustrated in Figure 31. The interest point was the same as the strain
gauge points during the experimental testing.

Figure 32: Longitudinal Strains: Outer Section - Ducati Rim

43
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

The points of interest were evaluated using sampling elements as highlighted in

Figure 22 .The sampling elements assisted in zoning into the strain gauged zones.
The general trend with respect to the longitudinal and transverse strains on the
spoke is highlighted in Figure 32 to Figure 35. The outer section of the spoke was in
tension, while the inner section was in compression. The strains are higher in the
hub region at the bolt attachment holes in line with the force application. The strains
in the hub region are expected to be high, due to the absence of the aluminium hub
in the FEA analysis. The results in the hub region were overlooked as they are not
modelled accurately. The hub results do not need to be accurate as the focus point
is on the spoke. The reason the results in the spokes are still valid was that the
overall constraints are still the same.

Figure 33: Longitudinal Strains: Inner Section - Ducati Rim

Figure 32 and Figure 33 illustrate the longitudinal strains on the Ducati rim. The
longitudinal strains on the outer and inner spoke are 164 µε and -113 µε respectively.
This illustrates that the outer spoke experiences the largest strain in the fibre
direction.

44
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The transverse strain of the Ducati rim is illustrated in Figure 34 and Figure 35. The
transverse strains of the outer and inner spoke are -582 µε and 454 µε respectively.
The transverse strains are smaller than the longitudinal strains. The in-plane shear
strains of the Ducati rim are illustrated in Figure 36 and Figure 37. The in-plane
shear strains for the outer and inner spoke are 328 µε and -662 µε respectively. The
strains obtained were used to calculate the longitudinal directional strains that can be
compared to the experimental data. The strains were obtained using equation 2.4.
The results are illustrated in Table 5.

Figure 34: Transverse Strain: Outer Section - Ducati Rim

45
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 35: Transverse Strains: Inner Section- Ducati Rim

Figure 36: In Plane Shear Strain: Outer Section - Ducati Rim

46
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 37: In Plane Shear Strain: Inner Section - Ducati Rim

The first step was to extract the relevant strain data (longitudinal, transverse and in-
plane shear strain) from the FEA model. The outputs from the FEA model were
utilised to calculate the longitudinal directional strains using eq. 2.4. The equation
was utilised by setting a local coordinate system for the strains experienced on the
inner and outer spoke. This was similar to the setup in Figure 7. The FEA results
were utilised in step 3 to perform the comparison with the experimental results as
illustrated in Table 5.

The FEA strains in the 45 °- direction have an error of 6.64% and 13.78% for the
inner spoke and the outer spoke respectively. The FEA strains in the 0 °- direction
have an error of -22.12% and -6.10% for the inner spoke and the outer spoke
respectively. Therefore the results are within a 25% accuracy range, thereby
indicating a reasonable correlation of the FEA and experimental data. Similar results
were obtained in the research report by Smith, B. which evaluates a carbon fibre
swingarm design [35]. The difference between the FEA and experimental results was
mainly due to the modelling difficulty of the structure, with reference to ply overlay
[35].

47
 Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 5: Comparison of FEA and Experimental Strains

FEA INPUTS
STEP 1

Outer (Front) Spoke Inner (Back) Spoke

ε_1 164 µε -113 µε
ε_2 -582 µε 454 µε
ε_12 328 µε -662 µε

STRAIN OUTPUTS (EQ. 2.4)

STEP 2

Outer (Front) Spoke Inner (Back) Spoke

ε_0 164 µε -113 µε
ε_45 295 µε -489 µε

CORRELATION OF RESULTS
Inner Inner Outer Outer
STEP 3

Spoke Spoke Spoke Spoke

(FEA) (Exp) Error (%) (FEA) (Exp) Error (%)
ε_0 -113 µε -138 µε -22.12% 164 µε 174 µε -6.10%
ε_45 -489 µε -465 µε 6.64% 295 µε 254 µε 13.78%

6.2.2 Maximum strain: failure criterion

The maximum strain failure criterion in the ACP platform was used to predict whether
fibre, matrix or in-plane shear failure will probably occur first [37]. The reserve factor
(RF) is a measure of the margin of failure or factor of safety [37]. The equation of the
reserve factor is indicated in equation 6.1. Therefore a RF, which is greater than one
indicates a low probability of failure.

𝑅𝐹 × 𝐹𝑎𝑝𝑝𝑙𝑖𝑒𝑑 = 𝐹𝑓 …………………………………………………………………. (Eq. 6.1)

Where:

RF Reserve Factor

𝐹𝑎𝑝𝑝𝑙𝑖𝑒𝑑 Applied load

𝐹𝑓 Failure load

48
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
In this report the inverse reserve factor (IRF) or inverse factor of safety was used to
evaluate the failure margin as illustrated in equation 6.2. This approach was more
practical in that the non-failure values of the IRF range from zero to one and the
failure values are above one [37].

1
𝐼𝑅𝐹 = 𝑅𝐹………………………………………………………………………. (Eq. 6.2)

The longitudinal, transverse and in-plane shear strains were analysed. In ACP the
maximum failure criterion is a ratio between the actual strains to the failure strains
that are compared in the ply principal coordinate system [37]. The maximum failure
criterion is illustrated in equation 6.3.

𝜀1 𝜀2 𝜀12
𝑓𝑚𝑎𝑥 = max(| 𝑊1 𝜀 | , |𝑊2 𝜀 | , |𝑊12 𝜀 |)………………………………. (Eq. 6.3)
1 𝑚𝑎𝑥 2 𝑚𝑎𝑥 12 𝑚𝑎𝑥

Where:

𝑓𝑚𝑎𝑥 Maximum failure criterion

𝜀1 Longitudinal strain

𝜀2 Transverse strain

𝜀12 In-plane shear strain

𝜀1 𝑚𝑎𝑥 Longitudinal failure strain

𝜀2 𝑚𝑎𝑥 Transverse failure strain

𝜀12 𝑚𝑎𝑥 In-plane shear failure strain

𝑊1 , 𝑊2 , 𝑊12 Weighting factors

In ACP, the failure criterion was defined as outlined in Figure 38. The longitudinal,
transverse and in-plane shear strains were analysed in conjunction with weighting
factors. The weighting (safety) factors were set at the value one for all the failure
mode criteria to ensure no bias is created for a failure mode. The advantage of the
weighting factors is that they can assist in creating focus on a specific failure mode
e.g. If the design may require a higher safety against fibre failure than delamination,
then the weighting factor of fibre failure is increased [37]. The results for the failure

49
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
modes are outlined in Figure 39 to Figure 47. It should be noted that the maximum
strain failure criterion (all modes) in the hub section would have failure regions
(values above one). This is due to the missing aluminium hub section. The results in
this section can be considered inaccurate; because the Ducati rim has a proven
service record. Rim Technology, Eurotype Test Centre and German TÜV have
tested the BST wheel and it has met all requirements with regards to road
certification [5].

Figure 38 : Defining the Maximum Failure Criterion

50
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 39: IRF - Longitudinal Strain Mode: Outer Section - Ducati Rim
The IRF in the longitudinal strain mode is illustrated in Figure 39 to Figure 41. The
IRF (Figure 39) in the spokes was 0.0909 on the outer section. The IRF (Figure 40,
Figure 41) in the spokes was 0.136 on the inner section.

Figure 40: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim

51
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 41: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim

The IRF in the transverse strain mode is illustrated in Figure 42 to Figure 44. The
IRF (Figure 42) in the spokes was 0.136 on the outer section. The IRF (Figure 43,
Figure 44) in the spokes was 0.0909 on the inner section.

Figure 42: IRF - Transverse Strain Mode: Outer Section - Ducati Rim

52
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 43: IRF - Transverse Strain Mode: Inner Section - Ducati Rim

Figure 44: IRF - Transverse Strain Mode: Inner Section - Ducati Rim

53
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 45: IRF- In Plane Shear Strain Mode: Outer Section - Ducati Rim

The IRF in the in-plane shear strain mode is illustrated in Figure 45 to Figure 47 .
The IRF (Figure 45) in the spokes was 0.0909 on the outer section. The IRF (Figure
46, Figure 47) in the spokes was 0.182 on the inner section.

Figure 46: Maximum Strain - In Plane Shear Strain Mode: Inner Section - Ducati Rim

54
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 47: IRF - In Plane Shear Strain Mode: Outer Section - Ducati Rim

In summary, the failure criterion or IRF indicates regions where the strain has
exceeded the failure strain of the material. The IRF in the hub section was above 1,
which was expected because the hub region was not modelled accurately. The hub
region was modelled without the aluminium hub. The IRF from the spoke to the rim
section was below 1. This indicates that the rim was structurally durable for its
function and the field performance of the rim has proven that. Section 6.2.2 indicated
that the spoke section had a higher probability of failure in the shearing (in plane
shear) mode, than matrix (transverse strain) and fibre (longitudinal strain) mode in
accordance with the maximum strain failure criterion. In chapter 7 the design of the
rim was evaluated to accommodate vehicle loads.

55
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 7: FEA – Passenger Car Rim

In accordance with the German TÜV wheel standard, the bending moment ranges
from 2 KNm to 70 KNm for the cornering fatigue test [38]. This standard ranges from
light vehicles to heavy vehicles. For the purpose of the research paper the analysis
was performed on a passenger car rim that experiences a 2000 Nm during the
dynamic cornering fatigue test. The test essentially caters for rims that will be
utilised for Category M1 motor vehicles as classified by Government Gazette 30917.
The gazette defines M1 motor vehicles as a carriage of passengers that has at least
four wheels, which can accommodate not more than 8 passengers including the
driver [38].

Table 6: Example of Bending Moment Calculation for the Vehicle

Bending Moment
Vehicle (kg) 2331 d (m) 0
4 passengers (kg) 400 𝜇 0.7
Luggage load (kg) 100 F 1.5
Total Mass (kg) 2831 r_dyn (m) 0.2286
F_r (N) 30 % load of total load on
worst loaded rim 8332 M_max (Nm) 2000

The 2000 Nm bending moment caters for vehicle mass limit of 2331 kg as illustrated
in Table 6. This was calculated using equation 2.1, in order to confirm that the
common sedan, hatch and SUV can utilise this rim design. The bending moment
was chosen because it structurally caters for rims of vehicles such as the Polo Vivo,
Ford Focus, Ford Kuga, Toyota Yaris, and Golf Gti etc.

The passenger car rim was constrained on the rim flange as illustrated in Figure 48.
A 2000 N remote force was applied 1 m away from hub as illustrated in Figure 48.
The force was attached to the bolt holes on the inner and outer hub. The results
outlining the design of a passenger car carbon fibre rim are illustrated below.

56
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 48: Load Case of Passenger Vehicle Carbon Fibre Rim

The first step was to expose the Ducati rim to vehicle loads, in order to evaluate the
strain patterns the wheel experiences before any modifications were done. The
cornering fatigue test setup was similar to the Ducati rim in chapter 6, but the remote
force applied was 2000 N instead of 700 N. The results are illustrated in section 7.1.

57
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
7.1 Ducati Rim Exposed to Passenger Vehicle Loads

7.1.1 Strains

The longitudinal strains of the Ducati rim are illustrated in Figure 49 and Figure 50.
The longitudinal strains in the spoke region were 2000 µε and -2640 µε for the outer
and inner spoke respectively.

Figure 49: Longitudinal Strains: Outer Section - Ducati Rim (2000 Nm)

Figure 50: Longitudinal Strains: Inner Section - Ducati Rim (2000 Nm)

58
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The transverse strains of the Ducati rim are illustrated in Figure 51and Figure 52.
The transverse strains in the spoke region were 1210 µε and -3530 µε for the outer
and inner spoke respectively.

Figure 51: Transverse Strains: Outer Section - Ducati Rim (2000 Nm)

Figure 52: Transverse Strains: Inner Section - Ducati Rim (2000 Nm)
The in planes shear strains of the Ducati rim are illustrated in Figure 53 and Figure
54. The in planes shear strains in the spoke region were 3880 µε and -4780 µε for
the outer and inner spoke respectively.

59
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 53: In Plane Shear Strain: Outer Section - Ducati Rim (2000 Nm)

Figure 54: In Plane Shear Strain: Inner Section - Ducati Rim (2000 Nm)
The strain patterns were similar to those described in section 6.2.1, but higher in
comparison, due to the larger bending moment. The strains were generally higher in
the transition zones between the hub, spokes and the rim. The strain patterns
indicate that strain reduction was required in the hub, spokes, and rim. Strain
reduction was especially required in the interface zone between the components.

60
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The design of the passenger vehicle carbon fibre rim was iterative to obtain a
feasible design. The design of the passenger vehicle is outlined in section 7.2 and
7.3. In these sections the FEA methodology is outlined, which entailed model
preparation, material design and the loading for the rim. These aspects are similar to
the Ducati rim setup and the differences will be highlighted to avoid duplication.

The strain results were crucial and assisted in understanding the Ducati rim at higher
loads. The results illustrate the strain patterns for the longitudinal, transverse and in
plane shear strains. The results can be utilised to compare the passenger rim to the
Ducati rim at lower (700 Nm) and higher (2000 Nm) loads. This assisted in
quantifying the improvement in the design of the passenger rim.

7.1.2 Maximum strain failure criterion

The IRF in the longitudinal strain mode is illustrated in Figure 55 and Figure 56. The
IRF (Figure 55) in the spokes was 0.409 on the outer section. The IRF (Figure 56) in
the spokes was 0.409 on the inner section.

Figure 55: IRF- Longitudinal Strain Mode: Outer Section - Ducati Rim (2000 Nm)

61
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 56: IRF - Longitudinal Strain Mode: Inner Section - Ducati Rim (2000 Nm)
The IRF in the transverse strain mode is illustrated in Figure 57 and Figure 58. The
IRF (Figure 57) in the spokes was 0.636 on the outer section. The IRF (Figure 58) in
the spokes was 0.409 on the inner section.

Figure 57: IRF - Transverse Strain Mode: Outer Section - Ducati Rim (2000 Nm)

62
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 58: IRF- Transverse Strain Mode: Inner Section - Ducati Rim (2000 Nm)
The IRF in the in plane shear strain mode is illustrated in Figure 59 and Figure 60.
The IRF (Figure 59) in the spokes was 0.636 on the outer section. The IRF (Figure
60) in the spokes was 0.682 on the inner section.

Figure 59: IRF - In Plane Shear Strain Mode: Outer Section - Ducati Rim (2000 Nm)

63
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 60: IRF - In Plane Shear Strain Mode: Inner Section - Ducati Rim (2000 Nm)
The IRF in the hub section was above 1, which was expected because the hub
region was not modelled accurately. Section 7.1.2 indicated that the spoke section
had a higher probability of failure in the shearing (in plane shear) mode, than matrix
(transverse strain) and fibre (longitudinal strain) mode in accordance with the
maximum strain failure criterion. The IRF will be compared to passenger rim results
in section 7.3.

7.2 Model Preparation

The existing 1098 Ducati surface CAD model was the reference point for the vehicle
rim design. The project schematic as illustrated in Figure 13 was the flow path
followed in the design. The existing material as illustrated in Table 3 was used in the
new design. The process of the design was iterative, with respect to segmentation,
stacking sequence and meshing. These items are discussed further in the report.

64
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
7.2.1 Segmentation

Segmentation of the rim was essential in the build-up of the passenger car rim. The
rim was segmented using named selections. This allowed for geometric definition of
the parts on the rim. The named selections (segments) illustrated in Figure 15 to
Figure 18 were still valid for the passenger car rim, except for the inner hub and
outer hub. The alterations to the hub segments are illustrated in Figure 61. The hub
was extended into the spoke section to ensure that there was reduction of strain in
the hub to spoke transition zone. The applicable coordinate systems are illustrated in
Figure 19 and Figure 20 for the respective segments as discussed in [Link].

Figure 61: Passenger Car Carbon Fibre Rim Hub Segments

7.2.2 Stacking sequence

The stacking sequence is illustrated in Table 7 and was very similar to the original
1098 Ducati rim layup. The stacking sequence was modelled in the modelling plys of
Pre ACP. The stacking sequence resulted in a thickness variation as illustrated in
Figure 62 and Figure 63. The increased thickness in the transition zones between
the hubs, spokes and the rim ensured the passenger rim had the required structural
integrity. The thickness variation of the laminate; varies from 6.2 mm thick at the hub

to 4.04 mm thick at the spokes, to 2.16 mm thick at the rim section (beads excluded).

65
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 7: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Hub Section

Segment Name Ply Type Ply Phase 1 :

Orientation Quantity
1.T300 woven 0º & ±45º Three

1.T300 woven 0º & ±45º Three

1.T700 Woven 1. 0º & ±45º 1. Nine per

hub

1.T700 Woven 1. 0º & ±45º 1. Nine per

hub

66
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 8: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Spoke Section

Segment Name Ply Type Ply Phase 1 :

Orientation Quantity
1.T300 woven 1. ±45º [Link] per
2. T700 2. UD spoke
unidirectional 3. ±45º [Link] per
3. T300 woven spoke
[Link] per
spoke

1.T300 woven 1. ±45º [Link] per

2. T700 2. UD spoke
unidirectional 3. ±45º [Link] per
3. T300 woven spoke
[Link] per
spoke

Intensifiers 1.T300 Woven 1.±45º 1. Rolled three

times around
intensifiers

67
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 9: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Rim Section

Segment Name Ply Type Ply Phase 1 :

Orientation Quantity
2.T300 woven ±45º Three arc
between
spokes on the
inner and
outer section
of the rim

1.T700 woven 1.±45º 1. Rolled nine

times around
the rim.

Plies for bead formation

excluded in layup, please refer
to original 1098 Ducati rim
design.

68
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 10: Stacking Sequence of the Passenger Car Carbon Fibre Rim: Rim Re-enforcement

Segment Name Ply Type Ply Phase 1 :

Orientation Quantity
1.T300 woven 0/90º Three per
spoke on
inner and
outer section
of the rim

1.T300 woven 1. ±45º 1. Five layers

between
spokes.

69
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 62: Thickness Variation: Passenger Car Rim

Figure 63: Thickness Variation: Outer Section- Passenger Car Rim

70
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
7.2.3 Meshing

Shell 181 elements were utilised in analysing the layers of the composite passenger
car rim. The mesh was refined as illustrated in Figure 64 in order to evaluate the
mesh sizing. The refined spoke was in line with the force application and the strain
results extracted from the spoke were compared to the Ducati rim strain gauge
results.

Figure 64: Refinement of the Passenger Car Mesh

The mesh sizing relevance centre and smoothing was set to a medium and the
minimum size was 0.5795 mm. There were 69091 nodes and 69069 elements. The
refinement of the illustrated sections in Figure 64 was such that the mesh was two
times smaller than the rest of the rim geometry mesh. The refinement of the mesh
caused a small change in the results as illustrated in Table 4.

Table 11: Evaluation of Mesh Sizing (Passenger Car Rim)

Medium Size
Strain Mesh Refined Mesh Difference
Longitudinal- 1Axis (µε) 688.11 688.39 0.04%
Transverse- 2 Axis (µε) 16.057 16.065 0.05%
In Plane Shear -12 Axis
(µε) -126.45 -126.501 0.04%

The discretization error and element size analysis was conducted on the refined
spoke region. A strain probe was placed to evaluate convergence of the results. The
mesh was initially medium size and it was later refined, in order to compare the two

71
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
stages. It was evident that the refinement of the mesh caused a small change in the
results; indicating that the sizing of the mesh was appropriate. The FEA results of
the passenger car carbon fibre rim are illustrated in section 7.3.

7.3 Result Evaluation: Passenger Vehicle Rim

The strain flow patterns for the passenger car rim are similar to the Ducati rim when
exposed to the motorcycle loads. The strain results were considerably reduced due
to the new layup in comparison to the Ducati rim exposed to vehicle loads. The
longitudinal, transverse and in plane shear strain of the spokes are essential to
evaluate the structural integrity of the passenger car rim as illustrated in section
7.2.1.

7.3.1 Strains

The longitudinal, transverse and in-plane shear strain patterns are similar to the
Ducati rim. The longitudinal strain for the passenger car was illustrated in Figure 65
and Figure 66. The longitudinal strains on the outer and inner spoke were 335 µε
and -466 µε respectively. This illustrates that the outer spoke experiences the
largest strain in the fibre direction.

The transverse strain of the passenger car is illustrated in Figure 67 and Figure 68.
The transverse strains of the outer and inner spoke were 189 µε and -532 µε
respectively. The in-plane shear strain of the Ducati rim is illustrated in Figure 69 and
Figure 70. The in-plane shear strains for the outer and inner spoke were 460 µε and
-650 µε respectively. The strains obtained were used to calculate the longitudinal
directional strains that can be compared to the FEA results of the Ducati rim. The
strains were obtained using equation 2.4. The results are illustrated in Table 12.

72
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 65: Longitudinal Strain: Outer Section - Passenger Car Rim

Figure 66: Longitudinal Strain: Inner Section - Passenger Car Rim

73
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 67: Transverse Strain: Outer Section - Passenger Car Rim

Figure 68: Transverse Strain: Inner Section - Passenger Car Rim

74
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 69: In-Plane Shear Strain: Outer Section - Passenger Car Rim

Figure 70: In-Plane Strain: Inner Section - Passenger Car Rim

75
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Table 12: Comparison of the FEA strain of Passenger Car and Ducati Rim

FEA INPUTS
STEP 1

Inner (Back) Spoke Outer (Front) Spoke

ε_1 -466 µε 335 µε
ε_2 -532 µε 189 µε
ε_12 -630 µε 460 µε

STRAIN OUTPUTS (EQ. 2.4)

STEP 2

Inner Spoke Outer Spoke

ε_0 -466 µε 335 µε
ε_45 -935 µε 639 µε

COMPARISON OF FEA RESULTS

Inner Outer
STEP 3

Spoke Inner Spoke Outer

(Ducati- Spoke Diff. (Ducati- Spoke Diff.
FEA) (Car- FEA) (%) FEA) (Car- FEA) (%)
ε_0 -113 µε -466 µε 76 % 164 µε 335 µε 51 %
ε_45 -498 µε -935 µε 47 % 295 µε 638 µε 53 %

Table 12 illustrates that the passenger rim generally experienced higher strain in
comparison to the Ducati rim exposed to 700 Nm. Table 13 illustrates that the
passenger rim resulted in an average 85% improvement in the strains when
compared to the Ducati rim when exposed to 2000 Nm. This illustrates that the
passenger rim was able to handle the higher vehicle loads, given that the strains are
still generally low. This was substantiated by the maximum strain failure criterion
evaluation in section 7.3.2.

Table 13: Comparison of the Ducati (2000 Nm) and Passenger Rim Strains

Ducati Rim Passenger Car

(2000Nm) (2000Nm) Difference
Inner Outer Inner Outer Inner Outer
Strains Spoke Spoke Spoke Spoke Spoke Spoke

ε_1 (µε) -2640 2000 -466 335 82% 83%

ε_2 (µε) -3530 1210 -532 189 85% 84%

ε_12
(µε) -4780 3880 -630 460 87% 88%

76
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
7.3.2 Maximum strain failure criterion

The maximum strain evaluation process of the passenger rim was similar to the
process followed in section 6.2.2. The value for the weighting factor for all the failure
modes was one. This was to ensure that the failure modes could be compared fairly.
The IRF in the longitudinal strain mode is illustrated in Figure 71 to Figure 73. The
IRF (Figure 71) in the spokes was 0.136 on the outer section. The IRF (Figure 72,
Figure 73) in the spokes was 0.136 respectively on the inner section.

Figure 71: Maximum Strain Failure Criterion - Longitudinal Strain Mode: Outer
Section - Passenger Rim

77
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 72: IRF - Longitudinal Strain Mode: Inner Section - Passenger Rim

Figure 73: IRF - Longitudinal Strain Mode: Inner Section - Passenger Rim

78
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 74: IRF - Transverse Strain Mode: Outer Section - Passenger Rim
The IRF in the transverse strain mode is illustrated in Figure 74 to Figure 76 . The
IRF (Figure 74) in the spokes was 0.227 on the outer section. The IRF (Figure 75,
Figure 76) in the spokes was 0.136 on the inner section.

Figure 75: IRF - Transverse Strain Mode: Inner Section - Passenger Rim

79
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 76: IRF - Transverse Strain Mode: Inner Section - Passenger Rim
The IRF in the in-plane shear strain mode is illustrated in Figure 77 to Figure 79.
The IRF (Figure 77) in the spokes was 0.273 on the outer section. The IRF (Figure
78, Figure 79) in the spokes was 0.318 on the inner section.

Figure 77: IRF - In Plane Shear Strain Mode: Outer Section - Passenger Rim

80
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 78: IRF - In Plane Shear Strain Mode: Inner Section - Passenger Rim

Figure 79: IRF - In Plane Shear Strain Mode: Inner Section - Passenger Rim
In summary the inverse reserve factor in the hub section was above 1, which was
expected because the hub region was not modelled accurately. The hub region was
modelled without the aluminium hub. The IRF from the spoke to the rim section was
below 1 and relatively low. This indicates a low risk of static failure. The maximum
strain patterns are similar to the Ducati rim exposed to 700 Nm. Table 14 illustrated
an average 60% improvement in the IRF of the passenger car rim in comparison to
the Ducati rim exposed to 2000 Nm. Section 7.2.2 indicated that the spoke section
had a higher probability of failure in the shearing (in plane shear) mode, than matrix

81
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
(transverse strain) and fibre (longitudinal strain) mode in accordance with the
maximum strain failure criterion.

Table 14: Comparison of the Ducati (2000 Nm) and Passenger Rim IRF

Ducati Rim Passenger Car

(2000Nm) (2000Nm) Difference
Inner Outer Inner Outer Inner Outer
IRF Spoke Spoke Spoke Spoke Spoke Spoke
ε_1
Mode 0.409 0.409 0.136 0.136 67% 67%
ε_2
Mode 0.409 0.636 0.136 0.227 67% 64%
ε_12
Mode 0.636 0.682 0.318 0.273 50% 60%

82
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Chapter 8: Discussion and Conclusion

8.1 Discussion

In the process of designing a carbon fibre passenger car rim, it was essential to
correlate the FEA and the experimental strain results of the 1098 Ducati rim as
illustrated in Table 5. This assisted in validating the finite element technique of the
carbon fibre passenger car rim. The Ducati rim was used as a benchmark in that it is
a proven product and meets all the necessary specifications. The Ducati rim was
experimentally tested in accordance with the SAE J328 cornering fatigue test
specification. The rim flange was fixed and constrained while a force was applied to
the hub. The force was applied 1 m away from the hub, thereby resulting in a
bending moment and simulating the cornering fatigue test.

The cornering fatigue test on the Ducati rim was conducted on the 09/06/2010 by
Kienhöfer. The results of the experimental analysis illustrate:

- The zero static voltage results obtained were stable, and indicated a low drift.
- Channel 03 (back face of spoke 45 degrees) measured the highest strains of
448.9 µε and - 465.1 µε during tension and compression. Channel 01 (front
face of spoke 45 degrees) experienced strains of 254.2 µε and -245.8 µε
during tension and compression. The highest strains were along the 45
degrees direction on the spokes.
- Channel 02 (back face of spoke 0 degrees) measured the lowest strains at
120.4 µε and – 137.6 µε respectively during tension and compression.
Channel 0 (front face of spoke 0 degrees) experienced strains of 173.8 µε and
-165.2 µε respectively during tension and compression. This indicates that the
spokes are well re-enforced in the 0 degrees direction along the spoke.
- The strain gauges experienced their highest strains, when the force
application direction aligned with the orientation of the strain gauges
(instrumented spoke). This was evident in the phase differences between the
channels in Figure 10.

83
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
- Figure 10 illustrates that the strain gauges experienced two peaks. The peaks
illustrate compression and tensile phases during the rotation of the rim. A
sinusoidal relationship was visible between the strain and time as the wheel
was rotated. These finding are similar to the findings of Wang et al. [6], with
respect to the strain, time and rotation of the wheel.
- In an ideal situation, the strain gauge results would be utilised to calculate the
longitudinal, transverse, and the in-plane shear strain. The absence of a third
strain result on the front and back surface limited the strain calculations
(longitudinal, transverse and the in plane shear strains). Therefore the strain
gauge results were compared to the FEA strains using equation 2.4.

The experimental strain gauge results were used to validate the FEA model. The
FEA model of the 1098 Ducati was developed using ACP. The CAD model was
prepared by converting the solid model into a surface model. The surface model
allowed for material design of the model in the FEA environment. The material
design was done in ACP (pre), which entailed the segmentation, material properties,
orientation and the stacking sequence of the plies. The results in this section were
instrumental in understanding the strains the rim was exposed to. The aspects are
highlighted below:

Figure 80: Example of Strain Variation

- The correlation between the experimental and FEA results was within 25%.
The longitudinal, transverse, in plane shear strain on spoke were utilised to

84
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
calculate the strain in the 0 and 45 degree direction as illustrated Table 5.
The 25% correlation was deemed acceptable.
- The hub section experienced the highest strain. This was expected as this
was the attachment point for the load application. The high values around the
bolt holes, was due to the stress raisers (concentration) created by removed
material (attachments holes). The fact that the aluminium hub was not part of
the analysis was noted. The results in the hub region were overlooked as they
were not modelled accurately. The hub results do not need to be accurate as
the focus point was the spoke. The reason the results in the spokes are still
valid was that the overall constraints are still the same.
- The spokes experienced higher strain at the transition zone C from the hub to
spokes as illustrated in Figure 80. Section C includes the spoke to spokes
sections. The load was applied directly above spoke B, but spoke D
experienced the highest strains along the spokes, which illustrated the
moment due to the force. The strains were higher on the curvature corners of
the spokes, and lower on section A of the spokes.
- The strains were generally higher at the point of the force application,
transition between segmentations and where there was a change in fibre
direction.
- The maximum strain failure criterion indicated that the Ducati rim was durable.
The inverse reserve factor were relatively low (below 1) on the spokes and the
rim. The inverse reserve factor in the hub section was above 1, which was
expected because the hub region was not modelled accurately. The hub
region was modelled without the aluminium hub and the results should be
overlooked.
- Section 6.2.2 indicated that the spoke section had a higher probability of
failure in the shearing (in plane shear) mode, than matrix (transverse strain)
and fibre (longitudinal strain) mode in accordance with the maximum strain
failure criterion. This indicates the rim is highly unlikely to experience static
failure.

The Ducati rim was exposed to 2000 Nm in order to evaluate the strains
(longitudinal, transverse and in plane shear). This assisted in quantifying the

85
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
improvement in the passenger rim design. The passenger rim was designed for a
2000 Nm bending moment. The design process for the stacking sequence and ply
orientation was iterative. The final layup process is detailed in Table 7. The
thickness variation (Figure 62 and Figure 63) of the rim; varies from 6.2mm thick
hub to a 4.04 mm thick spokes and a rim section of 2.16 mm (beads excluded). The
thickness change in comparison to the current rim is illustrated in Table 15. The
results are discussed below:

Table 15: Thickness Comparison of the Rims

Thickness Comparison
Ducati Passenger Car %
Rim Rim Change
Hub (mm) 4.1 6.2 51.22%
Spoke (mm) 2.5 4.04 61.60%
Rim Section
(mm) 1.8 2.16 20.00%

- The strains in the 0 and 45 degree direction on the spoke were calculated
using equation 2.4. Table 12 illustrates that the passenger rim generally
experienced higher strains in comparison to the Ducati rim exposed to 700
Nm. Table 13 illustrates that the passenger rim resulted in an average 85 %
improvement in the strains when compared to the Ducati rim when exposed to
2000 Nm.
- The maximum strain failure criterion of the passenger car rim is indicated in
section 7.2.2. The inverse reserve factor in the hub section was above 1,
which was expected because the hub region was not modelled accurately.
The hub region was modelled without the aluminium hub. The passenger car
rim experienced higher strains than the Ducati rim exposed to 700 Nm, but
the inverse reserve factor from the spoke to the rim section was below 1. This
indicates low risk of failure. The inverse reserve factor patterns are similar to
the Ducati rim. Since the Ducati rim was the benchmark, this proves that the
passenger car rim will not fail in service.
- Section 7.2.2 indicated that the passenger rim spoke section had a higher
probability of failure in the shearing (in plane shear) mode, than matrix
(transverse strain) and fibre (longitudinal strain) mode in accordance with the

86
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
maximum strain failure criterion. Table 14 illustrated an average 60%
improvement in the inverse reserve factor of the passenger car rim in
comparison to the Ducati rim exposed to 2000 Nm.

The design utilises the available shapes and material from the original design. The
designed carbon fibre passenger car rim weighs 3.56 kg in comparison to the
average weight of 17” inch aluminium rims of 9.42 kg [39]. This illustrates a weight
saving of 62. 2%.

87
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
8.2 Conclusion and Recommendations

In conclusion a carbon fibre passenger car rim can be designed and manufactured.
The main steps that were carried out to substantiate the feasible use of carbon fibre
in a passenger car rim are detailed below:

- The correlation between the experimental and FEA was within 25%. This
validated the finite element technique of the carbon fibre passenger car rim.
- The Ducati and passenger car rim experienced the highest strains in the hub
region. This was expected as this was the attachment point for the load
application. The high values around the bolt holes, were due to the stress
raisers (concentration) created by the removed material (attachments holes).
Since the hub was not modelled accurately, the results in this region were
neglected. The passenger car rim spoke experienced higher strains when
compared to the Ducati rim.
- The strain patterns in the transition zone and regions where there was a
change in fibre direction were generally higher.
- The maximum strain failure criterion indicated that the Ducati and passenger
car rim have a low probability of static failure.
- In comparison to aluminium rims of the equivalent size, a weight saving of
62.2% should be expected from the carbon fibre passenger car rim.
- Therefore a passenger car rim is feasible based on the principles listed
above.

Further work can be done to:

- Experimentally measure of the strains on the Ducati rim, using 45 degree

rosettes to measure all three surface strain components.
- Redesign the aluminium hub attachment for the passenger car carbon fibre
rim.
- Evaluate and reduce the probability of the carbon fibre degrading due to high
temperatures due to the heating on the brakes.
- Build the designed passenger rim and test it experimentally

88
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

References

1. Muthuraj, R., Badrinaryanan, R. and Sundararajan, T. (2011)

“Improvement in the Wheel Design Using Realistic Loading Conditions-
FEA and Experimental Stress Comparison.” SAE International. SAE
Papers 2011-28-0106.
2. Jacobs, H. (2002), “The Fatigue and Tensile Properties of A356 Aluminium
Alloy Wheel in Various Post Cast Conditions”, Department of Mechanical
Engineering Rand Afrikaans University, May 2002.
3. Raju, PR., Satyanarayana, B., Ramji, K., and Babu, KS. (2008) “Evaluation
of Fatigue Life of Aluminium Alloy Wheels Under Bending Loads.” Fatigue
and Fracture of Engineering Material and Structures, Department of
Mechanical Engineering, SRKR Engineering College, India.
4. Compere, AL., Griffith, WL. and Leitten, CF. (2001) “Improving the
Fundamental Properties of Line-Based Carbon Fibre for Transport
Applications.”, Oak Ridge National Laboratory.
5. BlackStone Tek, [Link], online, last cited 01/01/13.
6. Sali, A., Turner, G. and Kienhöfer, F. (2010) “The Novel Use of Carbon
Fibre in Designing an Automotive Car Rim.” Seventh South African
Conference on Computational and Applied Mechanics (SACAM 2010).
7. Grech-Cumbo, U.M., (2007) “Feasibility of a Carbon Fibre Racing Car
Wheel.” University of Witwatersrand
8. Kienhöfer, F.,(2010) “Experimental Results: Strain gauge results of 1098
Ducati Rim.”, The University of the Witwatersrand.
9. Radulescu, AV., Cananau, S. and Radulescu, L. (2012) “Mechanical
Testing Methods Concerning the Stress Analysis For a Vehicle Rim.”
Volume 2, pp. 33-39, Romania, University Politehnica Burcharest.
10. Kinstler, J. (2005) “The Science and Methodology of SAE Wheel Fatigue
Test Specifications.” SAE International. SAE Papers 2005-01-1826.
11. Mohammed, K., Vinothkumar, S., Sabarinathan, S., AnilKumar, N. (2011)
“Simulation and Test Correlation of Wheel Impact.” SAE International.
SAE Papers 2011-28-0129.

89
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
12. Wang, X. and Zhang, X. (2010) “Simulation of Dynamic Cornering Fatigue
Test of a Steel Passenger Car Wheel.” International Journal of Fatigue 32,
pp. 434–442, Elsevier.
13. Society of Automotive Engineers. (2001) “Wheels-Passenger Car and
Light Truck Performance Requirements and Test Procedures.” SAE
International, SAE Handbook J328.

14. Karandikar, HM. and Fuchs, W. (1990) “Fatigue Life Prediction For Wheels
by Simulation of the Rotating Bending Test.” SAE International, SAE
Papers 900147.
15. Guo, M., Bhandarkar, R. and Lin, B. (2004) “Prediction of a Cast
Aluminium Wheel Using Finite Element Analysis.” SAE International, SAE
Papers 2004-01-1581.
16. Fischer, G. And Grubisic, VV. (1998) “Design Criteria and Durability
Approval of Wheels Hubs.” SAE International, SAE Papers 982840.
17. Kocabicak, U. And Firat, U. (2001) “Numerical Analysis of Wheel
Cornering Fatigue Tests.” Engineering Failure Analysis 8, pp. 339-354.
18. Riesner, M. and DeVries, RI. (1983) “Finite Element Analysis and
Structural Optimization of Vehicle Wheels.” SAE International. SAE
Papers 830133.
19. Wang, L., Chen, Y., Wang, C., and Wang, Q. (2010) “Fatigue Life Analysis
of Aluminium Wheels by Simulation of Rotary Fatigue Test.” Journal of
Mechanical Engineering 57,pp. 31-39, China, School of Mechanical
Engineering, Nanjing University of Science and Technology.
20. Statistics South Africa (2010) “Motor Trade Industry, 2009.” Statistical
Release P6301,
[Link]
21. Tsuboi, U. (1998) “Cultural Observations on the History of Wheels.” SAE
International. SAE Papers 980617.
22. Ashby, MF. (1999) “Material Selection in Material Design.” Chapter 6, pp.
119-120.
23. Baffet, G., Charara, A. and Lechner, D. (2009) “Estimation of Vehicle
Sideslip, Tire Force and Wheel Cornering Stiffness.” Volume 17, Issue 11,
pp. 1255-1264, Elsevier Science.

90
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
24. Pereira de Carvalho, C., Voorwald, HJ. and Lopes, CE. (2001)
“Automotive Wheels- An Approach for Structural Analysis and Fatigue Life
Prediction.” SAE International. SAE Papers 2001-01-4053.
25. TÜV (1998) “Directives on Testing of Special Wheels for Cars and Motor
Cycles.” Road Traffic Licensing Regulation, 30 STVZO
26. Euwa Standards (2006) “Test Requirements for truck steel wheels.” ES
3.11, EUWA.
27. Kaw, AK. (2005) “Mechanics of Composite Material.” Second edition,
London, CRC Taylor and Francis.
28. Efunda,[Link]
p_lamina_principal.cfm, online, last cited 01/09/13.
29. Efunda,[Link]
strain_gage_rosette.cfm, online, last cited 01/09/13.

30. Hsu, Y., Wang, S. And Liu, T. (2004) “Prediction of Fatigue Failures of
Aluminium Disc Wheels Using the Failure Probability Contour Based on
Historical Test Data.” Journal of the Chinese Institute of Industrial
Engineers, Vol. 21, No. 6, pp. 551-558.
31. Fleischhauer, M. and Koenen, J. (2008) “Non-Linear Simulation of a
lightweight Composite Rim.” JEC Magazine no. 45,
[Link].
32. Potter, K. (1997) “An Introduction to Composite Products.” First Edition,
pp. 163, London, Chapman and Hall.
33. Giek, JE. And Noll, WD. (1980), “Composite Wheels.” SAE International,
SAE Papers 800067.
34. Hexcel (2005),”Prepreg Technology.” Publication no. FGU 017b.
35. ANSYS,[Link]
online, last cited 01/01/13.
36. Smith, B., Kienhöfer, F., (2015) “A Carbon Fibre Swingarm Design.” R&D
Journal of South African Institution of Mechanical Engineering, pp. 1-11.
37. ANSYS (2013), “ANSYS Composite PrepPost User`s Guide.” ANSYS inc.
38. TÜV SÜD. “Wheel Testing.” Munich, Germany, [Link].
39. Wheel Weights, [Link], online, last cited 27/03/13.

91
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
40. SABS. “Motor Vehicles of Category M1.” Government Gazette 30917,
[Link].

41. Ashby, MF. (2009) “Materials- A Brief History.”, Philosophical Magazine

Letters, March 2009.
42. Kretschmer, J. (1988) “Composites in Automotive Applications- State of
the Art and Prospects.”, The Institute of Metals, March 1988.

43. Ashby, MF. Allwood, MJ., Gutowski, GT. and Worrell, E. (2010) “Material
Efficiency: A White Paper.”, Elsevier B.V, November 2010.
44. Reinforced plastics (2011) “Carbon Fibre Market Poised for Expansion.”
0034-3617/11Elsevier, pp 26-31, March 2011,
[Link].
45. Barbero, J. (2011) “Introduction to Composite Materials Design.” Second
Edition, pp 8- 226, New York, CRC Press.
46. Ashby, MF. and Jones, RH. (2002) “Engineering Materials 1: An
Introduction to their Properties and Applications.” Second Edition,
Butterworth Heinemann.
47. Poursartrip, A. and Beaumont, PWR. (1986) “The Fatigue Damage
Mechanics of a Carbon Fibre Composite Laminate: II- Life Prediction.”
Composite Science and Technology 25, pp 283-299.

48. Giger, M. and Ermanni, P. (2012) “Development of CFRP Motorcycle Rims

Using Evolutionary Algorithms.”, CH-8092 Zurich, Switzerland,
[Link]
49. Hibbeler, RC. (2004) “Engineers Mechanic Dynamics.” Third edition, pp
377,378,576, Prentice Hall.

50. Zuo, ZH., Xie, YM. and Huang, X. (2011) “Reinventing the Wheel.” Journal
of Mechanical Design, Vol. 33, School of civil, Environmental and
Chemical, [Link].
51. Rao, K., Rastogi, S., Upadhya, H., Sanmani, A., Hakeem, SA. and
Narayana, KB. (2011) “Damage Tolerance Analysis of Aero Structural
Components.”, TATA Consultancy Services..

92
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
52. Stearns, C. (2000) “An Investigation of Stress and Displacement
Distribution in a Aluminium Alloy Automobile Rim.”, The Graduate Faculty
of the university of Akron.
53. Garret, TK. (2001) “The Motor Vehicle”, Reed educational and
Professional Publishing Ltd.
54. Roeseler, W., Sarh, B. and Kismarton, M. (2007) “Composite Structures:
The First 100 Years.” 16th International Conference on Composite
Materials, Kyoto Japan.
55. Hsu,Y. (2006) “Developing an Automated Design Modification System for
Aluminium Disk Wheel.” 10th International Conference on CSCW, China.
56. Hibbler, RC. (2005) “Mechanics of Material.” 6th Edition, pp. 585-642,
Prentice Hall

93
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Appendix A

94
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Micromechanics of composite material

In micromechanics a combination of two isotropic materials, is characterised as an

equivalent, homogenous, anisotropic material [45]. The stiffness of the indicated
material is described by the listed properties in Figure 81.

Figure 81: Stress Illustration on Cube [edited 45]

Longitudinal Modulus

The modulus of elasticity in the fibre direction is based on the assumptions [44]:

- Strains in the direction of the fibre are the same in the matrix and the fibre.

- Homogenisation of composite (RVE – representative volume element)

95
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________

Figure 82: Longitudinal Uniform Strain Loaded on RVE [edited 45]

The area of the RVE is defined by area of the matrix and fibres as illustrated in
Figure 82. The total area (A) of the RVE is the addition of the matrix Area (A m) and
fibre area (Af). A Load (P) applied to the total area will result in the elongation of the
RVE from its original state [25]. The strain (𝜖1) is the ratio of the change in length
(∆L) to the original length (L). The load applied can be characterised by equation
below. The stress (σ1) – strain (𝜖1) law in relation to matrix stress/elastic modulus (σm
/Em) and fibres stress/elastic (σf/Ef) are illustrated below [45]:

𝜎𝑓=𝜖1 𝐸𝑓 𝜎𝑚=𝜖1 𝐸𝑚 𝑃 = 𝜎𝑓 𝐴𝑓 + 𝜎𝑚 𝐴𝑚

The longitudinal modulus is the upper limit estimate modulus [30]. The fibre-
reinforced composite modulus cannot exceed the upper limit, since matrix strains are
generally greater than fibre strains [45]. Therefore the longitudinal modulus is
indicated below [45]:

𝑬𝟏 = 𝑬𝒇 𝑽𝒇 + 𝑬𝒎 𝑽𝒎 .................................................................................... (Eq. A1)

Where:

Ef Fibre modulus

E1 Longitudinal modulus

96
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Vf (Af/A) Fibre volume percentage (%)

Em Matrix modulus

Vm (Am/A) Matrix volume percentage (%).

σf Fibre stress

σm Matrix stress

Transverse Modulus

The modulus of elasticity in direction of the transverse fibres is based on

assumptions similar to longitudinal fibres. E3 is generally estimated and assumed to
be equivalent to the transverse modulus. In the transverse modulus the fibre strains
and matrix strains act over a portion of the RVE width (W) [45]. The fibre strains
𝜎2 𝜎2
( 𝜖𝑓 = 𝐸 ) and matrix strains ( 𝜖𝑚 = 𝐸 ) respectively act over Vf W and VmW. The
𝑓 𝑚

average strain (𝜖2 ) is the total elongation of the RVE as represented below:

𝜖2 𝑊 = 𝜖𝑓 𝑊𝑉𝑓 + 𝜖𝑚 𝑊𝑉𝑚

The transverse modulus is the lower limit estimate modulus [45]. Therefore the
transverse modulus is indicated below [45]:

𝟏 𝑽 𝑽
= 𝑬𝒇 + 𝑬𝒎 ................................................................................................... (Eq. A2)
𝑬𝟐 𝒇 𝒎

Where:

E2 Transverse modulus

𝜎2 Transverse stress

In-Plane Poisson`s Ratio

Poisons Ratio is the negative ratio of the transverse strain (𝜖𝑗 ) to the axial strain (𝜖𝑥 )
[45]. Equation A3 provides an approximation of the Poisson’s ratio and is governed
by the rule of mixtures as a function of fibre volume and matrix volume fraction [45].

𝒗𝟏𝟐 = 𝒗𝒇 𝑽𝒇 + 𝒗𝒎 𝑽𝒇 ........................................................................................ (Eq.A3)

97
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Where:

𝒗𝟏𝟐 In-plane Poisson`s ratio

𝒗𝒇 Fibre Poisson`s ratio

𝒗𝒎 Matrix Poisson`s ratio

In-Plane Shear Modulus

Figure 83 illustrates the deformed composite due to in-plane shear. Equation 8 is a

matrix dominated property in case of stiff fibres. Shear through the thickness (G 13) is
assumed to be equal to in-plane shear (G12) [45].

𝟏 𝟏 𝑽 𝑽
=𝑮 = 𝑮𝒇 + 𝑮𝒎 ............................................................................................ (Eq. A4)
𝑮𝟏𝟐 𝟏𝟑 𝒇 𝒎

Figure 83: Illustration of In-Plane Shear [edited 45]

Where:

Gf Fibre shear modulus

Gm Matrix shear modulus

G12 In-plane shear modulus

98
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
Intra-laminar Shear Modulus

Figure 84: Illustration of Intra-Lamina Shear [edited 45]

Figure 84 illustrates the intra-laminar shear (𝐺23 ) acting across the thickness of the
composite. This acts along the 3-2 plane as illustrated in equation A5. Therefore
understanding composite micromechanics is crucial in FEA.

𝜈𝑓 +𝜂4 (1−𝜈𝑓 )
𝐺23 = 𝐺𝑚 𝐺 ..................................................................................... (Eq. A5)
𝜂4 (1−𝜈𝑓 )+𝑉𝑓 𝑚
𝐺𝑓

Where:
𝐺
3 − 4𝜈𝑚 + 𝐺𝑚
𝑓
𝜂4 =
4(1 − 𝜈𝑚 )

Fatigue Failure

Fatigue analysis is performed to evaluate cyclic loading and the reduction of residual
strength of the composite as illustrated in Figure 85 [48]. The cornering fatigue test
requires that the composite wheel be evaluated for fatigue failure. The first step in
fatigue analysis is to determine whether it is fibre or matrix dominated failure [45].
The composites resistance to failure is dependent on the composite, tensile and in-
plane shear strength [32]. Carbon fibre usually experience drastic failure and glass
fibres experience a long period in reduction of residual strength before failure [32].
Aramid fibres are the intermediate case between glass and carbon fibres as
illustrated in Figure 85 [32].

99
Design of a Carbon Fibre Passenger Car Rim
___________________________________________________________________
The Weibull distribution is utilised to describe the variations in fatigue strengths [47].
Miners Rule can be utilised to predict fatigue life of composites under cyclic load, but
this approach tends to overestimates fatigue life [47]. The fatigue life (𝑁𝑓 ) is a
function of the fixed stress ratio (R), initial damage (𝐷𝑖 ), change in tensile strength
(∆𝜎) and terminal damage (𝐷𝑓 ) as illustrated in equation A6 and A7 [46].

𝑑𝐷
= 𝑓(∆𝜎, 𝑅, 𝐷)........................................................................................ (Eq.A6)
𝑑𝑁

𝐷𝑓 𝑑𝐷
𝑁𝑓 = ∫𝐷𝑖 ...................................................................................... (Eq. A7)
𝑓(∆𝜎,𝑅,𝐷)

Figure 85: Residual Strength Curves of the Composites [edited-32]

Fatigue failure models aim at mathematically describing failure modes such as

matrix micro-cracking and delamination growth rate, which are dependent on the
strain energy release rate [45]. The fitness function of the rim will be based on the
maximum strain failure criterion, which must remain below the limit value of 1 [45].
The maximum strain failure criterion distinguishes between compressive and tensile
strengths of a lamina [25]. The maximum strain failure evaluates if the maximum
allowable of the compression, tensile and shear strain is exceeded.

100