<span class="text_page_counter">Trang 1</span><div class="page_container" data-page="1">

VIETNAM NATIONAL UNIVERSITY HO CHI MINH CITY HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY

FACULTY OF MECHANICAL ENGINEERING

CAPSTONE PROJECT

DESIGN CONVERSION MID-DRIVE MOTOR KIT FOR MTB (MOUNTAIN BIKE) TO E-MTB

(ELECTRIC MOUNTAIN BIKE)

Student’s name: Nguyen Cong Hieu Student ID: 1952048

Instructor: MSc. Luong Thanh Nhat

Ho Chi Minh City, 2024

</div><span class="text_page_counter">Trang 3</span><div class="page_container" data-page="3">

No.: _____/ĐHBK – CK

<b>FINAL YEAR PROJECT PROPOSAL </b>

(This form must be appeared at the first page in the report of the final year project)

<b>Students’s name: Nguyen Cong Hieu Student’s ID: 1952048 Program: </b> Mechatronics Engineering <b>Class: CC20COD1 </b>

- Research overview of problems related to e-MTB:

+ Reasons why the market needs e-MTB, especially e-MTB conversion kits. + Explore torque control technologies for e-Bikes.

+ Explore torque measurement technologies for e-Bikes

+ Propose and analyze feasible options and select the appropriate solutions for whole conversion kit for MTB to e-MTB.

- Mechanical design

+ Design the gearbox for the motor. + Design the crankset system. - Electrical system design

+ Hall effect sensor circuit design for the motor.

+ Temperature sensor circuit design for the motor and MOTFETs. + Modify the design of the controller circuit of VESC Open Source. - Experimental

+ Assemble mechanical and electrical system. + Experimental evaluation of current controller. + Experimental evaluation of pedal assist ability. - Technical drawings 5, including:

+ 1 A0 drawing, about: Conceptual design + 1 A0 drawing, about: Mechanical design

</div><span class="text_page_counter">Trang 4</span><div class="page_container" data-page="4">

Faculty: Department Date of defense Evaluation grade: Archived place:

</div><span class="text_page_counter">Trang 5</span><div class="page_container" data-page="5">

First and foremost, we extend our deepest thanks to our parents and family members for their unwavering trust, encouragement, and support throughout our educational journey.

I wish to convey my sincere appreciation to MSc. Luong Thanh Nhat, who imparted not only the essential knowledge but also the professional demeanour required of an engineer. His patience and enthusiasm in pointing out deficiencies in technical drawings and practical machining techniques significantly broadened our understanding and enriched our experience for our future endeavours.

I am also profoundly grateful to VIEROBOT Co., Ltd. for their invaluable assistance in providing ideas, equipment, research space, and funding for our project. Special thanks go to the electrical, mechanical, and embedded programming engineers who supported us throughout the implementation process.

Lastly, I extend our gratitude to the teachers of the HCMUT for their instruction and assistance, enabling us to complete the Mechatronics Bachelor Program. I am also thankful to our dear friends for their companionship, support, and collaboration on various assignments and projects. I hope to have the opportunity to work together again in the future.

</div><span class="text_page_counter">Trang 6</span><div class="page_container" data-page="6">

1.1.1. Growing demand for sustainable transportation source. ... 1

1.1.2. The best solution for developing sustainable personal transportation. ... 1

1.1.3. Definition of e-Bikes and e-MTB ... 2

1.1.4. The development potential of e-MTB and the need for e-MTB conversion kits. ... 2

1.2. Literature review ... 3

1.2.1. Domestic research ... 3

1.2.2. International research ... 4

1.3. Necessity of product research and development. ... 5

1.4. Research and implementation range. ... 6

Criteria 1: Working in Off-Road condition & Require power... 6

Criteria 2: Pedal assistance ... 6

Criteria 3: Lower limit for maximum speed at crankset ... 7

1.5. Structure of the thesis ... 7

CHAPTER 2: SELECTION METHODS ... 8

2.1. Mapping selections ... 8

2.2. Priority selections ... 9

2.2.1. Position of actuator & working principal ... 9

2.2.2. Battery cover selection ... 12

</div><span class="text_page_counter">Trang 7</span><div class="page_container" data-page="7">

2.3.1. Torque sensor selection ... 21

2.3.2. Temperature sensor selection... 22

2.3.3. Rotor position sensor selection ... 22

2.3.4. Select ESC open source ... 23

2.3.5. Select BMS & Cell pins ... 24

CHAPTER 3: MECHANICAL DESIGN ... 25

3.1. Working principle ... 26

3.1.1. Method to measure torque from both legs ... 26

3.1.2. Method motor assist the system ... 26

3.2. Motor selection and transmission ratio ... 27

3.2.1. Calculate the power for the motor and motor selection ... 27

3.2.2. Transmission ratio distribution ... 29

3.2.3. Technical specification of the transmission ... 31

3.3. Chain drive design ... 31

3.4. Gearbox design ... 36

3.4.1. Transmission gears: sun gear and planetary gear. ... 36

3.4.2. Transmission gears: planetary gear and ring gear. ... 39

3.4.3. Shaft design ... 42

3.4.4. Key selections ... 46

3.4.5. Bearing selection ... 49

3.5. Crankset design ... 51

3.6. Holder motor design ... 51

CHAPTER 4: ELECTRICAL – ELECTRONIC DESIGN ... 54

4.1. Controller hardware module ... 54

4.2. Hall effect sensors module ... 56

4.1.1. Position arrangement for hall effect sensors ... 56

4.1.2. Schematic ... 56

4.3. Temperature sensors ... 59

4.4. Bluetooth module ... 60

CHAPTER 5: CONTROL – ALGORITHM DESIGN ... 61

5.1. Current reference generator ... 62

</div><span class="text_page_counter">Trang 8</span><div class="page_container" data-page="8">

5.2. Current controller ... 63

CHAPTER 6: EXPERIMENTAL RESULTS AND EVALUATION ... 67

6.1. Mechanical processing and assembly ... 67

6.1.1. Gearbox ... 67

6.1.2. Crankset ... 69

6.1.3. Overall assembly ... 71

6.2. Electronic processing and assembly ... 74

6.2.1. Controller and driver ... 74

6.2.2. Hall effect sensors ... 75

</div><span class="text_page_counter">Trang 9</span><div class="page_container" data-page="9">

<b>INDEX OF FIGURE </b>

Figure 1.1: Compare the carbon footprints of the different vehicles ... 1

Figure 1.2: Growth Trends & Forecasts of e-MTB market size (2024 - 2029) ... 3

Figure 1.3: Growth Trends & Forecasts of e-Bikes market size (2024 - 2029) ... 3

Figure 1.4: A prototype mid-drive conversion kit to e-Bikes in IndieGoGo ... 5

Figure 2.1: Priority selections ... 8

Figure 2.2: Secondary selections ... 8

Figure 2.3: Mid-drive motor e-Bikes... 9

Figure 2.4: Hub-drive motor e-Bikes ... 9

Figure 2.5: The way to measure torque of the right leg ... 10

Figure 2.6: Proposed principle diagram ... 11

Figure 2.7: The system working without assistance ... 12

Figure 2.8: The system working with assistance ... 12

Figure 2.9: Product drawing from the RYOBI manufacturer ... 13

Figure 2.10: How to measure the current of the phases in FOC ... 15

Figure 2.11: Error between measured current vector and desired current vector ... 15

Figure 2.12: Converting to the fixed coordinate ... 16

Figure 2.13: Converting to the synchronous rotation coordinate ... 16

Figure 2.14: States of converting 3-phase to controllable 2-phase in FOC ... 16

Figure 2.15: Controllers for id and iq ... 17

Figure 2.16: States of converting controlled 2-phase to 3-phase in FOC ... 17

Figure 2.17: Additional reluctance torque of Toyota/ Prius Hybrid THS II Motor [25] ... 18

Figure 2.18: Stationary frame state observer for a salient machine [26] ... 18

Figure 2.19: Product drawing from the manufacturer ... 20

Figure 3.1: Working principle & main blocks for mechanical design ... 25

Figure 3.2: How the system measures the torque of the left leg ... 26

Figure 3.3: How the system measures the torque of the right leg ... 26

Figure 3.4: How does the system separate the torque of the motor and the cyclist ... 27

Figure 3.5: How does the system assist the cyclist ... 27

Figure 3.6: Tab Design of Chain Design in Autodesk Inventor ... 32

Figure 3.7: Tab Calculation of Chain Design in Autodesk Inventor ... 33

Figure 3.8: Tab Selection Chain of Chain Design in Autodesk Inventor ... 34

Figure 3.9: Working principle of the planetary gearbox with fixing the ring gear ... 36

Figure 3.10: Tab Design of Gear Design (Sun + Planet gears) in Autodesk Inventor .. 36

Figure 3.11: Tab Calculation of Gear Design (Sun + Planet gears) in Autodesk Inventor ... 37

Figure 3.12: Tab Design of Gear Design (Ring + Planet gears) in Autodesk Inventor . 39Figure 3.13: Tab Calculation of Gear Design (Ring + Planet gears) in Autodesk Inventor ... 40

</div><span class="text_page_counter">Trang 10</span><div class="page_container" data-page="10">

Figure 3.14: Arrange the distance on the shaft ... 42

Figure 3.15: Tab Design of Shaft Design in Autodesk Inventor ... 43

Figure 3.16: Analyzing the force acting on the shaft ... 43

Figure 3.17: Tab Calculation of Shaft Design in Autodesk Inventor ... 44

Figure 3.18: Shear Force Graph, YZ Plane ... 44

Figure 3.19: Shear Force Graph, XZ Plane ... 45

Figure 3.20: Bending Moment Graph, YZ Plane ... 45

Figure 3.21: Bending Moment Graph, XZ Plane ... 45

Figure 3.22: Ideal Diameter of Shaft ... 46

Figure 3.23: Tab Design of Key (Chain Sprocket) in Autodesk Inventor ... 46

Figure 3.24: Tab Calculation of Key (Chain Sprocket) in Autodesk Inventor ... 47

Figure 3.25: Tab Design of Key (Chain Clutch) in Autodesk Inventor... 48

Figure 3.26: Tab Calculation of Key (Chain Clutch) in Autodesk Inventor ... 48

Figure 3.27: Tab Calculation of Bearing Selection in Autodesk Inventor ... 50

Figure 3.28: Put the force the chain system exerts on the shaft in Autodesk Fusion .... 52

Figure 3.29: Define the material for objects being analyzed ... 52

Figure 3.30: Displacement of the Motor after solving in Autodesk Fusion ... 53

Figure 3.31: Safety factor of the Motor after solving in Autodesk Fusion ... 53

Figure 4.1: Reference schematic design of Open source VESC ... 54

Figure 4.2: General schematic of the controller hardware ... 55

Figure 4.3: 3D images of the controller hardware ... 56

Figure 4.4: Hall effect sensors arrangement ... 56

Figure 4.5: Prototype schematic of the hall effect sensors module ... 57

Figure 4.6: Hall effect sensors module signal at the prototype version ... 58

Figure 4.7: Final schematic of the hall effect sensors module ... 58

Figure 4.8: Hall effect sensors module signal at the final version ... 59

Figure 4.9: Reference design for NTC sensors ... 59

Figure 4.10: Wiring diagram of temperature sensors inside of the ESC (left side) and motor (right side) ... 60

Figure 4.11: Schematic of the Bluetooth module ... 60

Figure 5.1: General block diagram of motor support ... 61

Figure 5.2: Block diagram of current control by the FOC ... 61

Figure 5.3: The current control system is divided into two blocks ... 64

Figure 5.4: Block diagram after prediction and reduction ... 64

Figure 5.5: Pole suppression plan in current control ... 65

Figure 5.6: Transfer function prediction results in VESC Tool ... 65

Figure 6.1: Some images about components of the gearbox are 3D printed ... 67

Figure 6.2: Some images after machining and assembly the gearbox ... 68

Figure 6.3: The prototype version of the crankset ... 69

Figure 6.4: Version 1 of the crankset with changing to block aluminium ... 70

</div><span class="text_page_counter">Trang 11</span><div class="page_container" data-page="11">

Figure 6.5: Freewheel damage when going off-road for over 60km ... 70

Figure 6.6: The crankset version 2 has replaced the freewheel with the bearing... 71

Figure 6.7: Several images after whole assembly ... 72

Figure 6.8: Several render image of mechanical system Version 2 ... 73

Figure 6.9: Moisture damage, particularly those subjected to significant loads and high temperatures ... 74

Figure 6.10: A layer of nylon is applied to the torque sensor to enhance its durability and resistance to moisture ... 74

Figure 6.11: Component replacement process ... 74

Figure 6.12: The PCB circuit after processing ... 75

Figure 6.13: Mounting the PCB into the stator of the motor ... 75

Figure 6.14: Sensor signal when they are placed in the wrong face angle ... 76

Figure 6.15: Sensor signal when they are far the rotor ... 77

Figure 6.16: The usable signal of hall effect sensors ... 77

Figure 6.17: Some temperature sensor installation locations ... 78

Figure 6.18: Reading the temperature signal ... 78

Figure 6.19: Some images about installation the Bluetooth module ... 79

Figure 6.20: The experimental setup for recalibrating the cadence speed ... 80

Figure 6.21: Packaging a battery ... 80

Figure 6.22: Balance the battery cells before packaging the battery... 81

Figure 6.23: Balance the battery cells after packaging the battery ... 81

Figure 6.24: Preparation for the current control experiment ... 82

Figure 6.25:Experiment result with the step inputs (8A to 19A, each step 1A) ... 83

Figure 6.26: Experiment result with the step input (0A to 10A, one step 10A) ... 83

Figure 6.27: Motor speed work no load at sensor mode ... 84

Figure 6.28: Motor speed work no load at sensorless mode ... 84

Figure 6.29: Motor speed work on the heavy load at sensorless mode ... 85

Figure 6.30: Motor speed work on the heavy load at sensor mode ... 85

Figure 6.31: The motor operates under heavy load in sensor mode when the ERPM is below 5000, and transitions to sensorless mode when the ERPM exceeds 5000. ... 86

Figure 6.32: Setup motor before the pedal-assist experiment ... 87

Figure 6.33: Logging data at Tour Mode ... 88

Figure 7.1: Total distance traveled under assistive mode ... 90

</div><span class="text_page_counter">Trang 12</span><div class="page_container" data-page="12">

<b>INDEX OF TABLE </b>

Table 1.1: Constraints for Criteria 1 in summary ... 6

Table 2.1: Comparison between Mid-drive and Hub-drive ... 9

Table 2.2: Comparison between current control methods ... 14

Table 2.3: Comparison between types of clutches in gearbox ... 19

Table 2.4: Comparison between types of clutches in crankset ... 20

Table 2.5: Comparison between types of torque sensors ... 21

Table 2.6: Specification of T13 torque sensor ... 22

Table 2.7: Comparison between types of rotation position sensors ... 22

Table 2.8: Comparison between two popular FOC open sources ... 24

Table 3.1: Specification of the motor ... 29

Table 3.2: Select transmission ratio for planetary gear box ... 30

Table 3.3: Transmission ratio distribution ... 31

Table 4.1: Specifications of the A3144EUA hall effect sensor ... 57

Table 6.1: Compare upgrades in two gearbox versions ... 68

Table 6.2: Compare upgrades in two crankset versions ... 71

Table 6.3: Compare other upgrades in two versions ... 71

</div><span class="text_page_counter">Trang 13</span><div class="page_container" data-page="13">

<b>CHAPTER 1: OVERVIEW 1.1. Problem statement </b>

1.1.1. Growing demand for sustainable transportation source.

Mobility plays a vital role in the current society that we live in. With the growing concern for preserving and sustaining it for future generations, mainly because of the increasing scarcity of natural resources and environmental concerns, protecting the environment poses a significant challenge to society and governments worldwide.

Owing to this, some international organizations worldwide are implementing strict criteria for vehicles. For instance, In January 2020, the European Union implemented Regulation (EU) 2019/631, setting CO2 emission performance standards for new passenger cars and vans. The average CO2 emissions from new passenger cars registered in Europe have decreased by 12% compared to the previous year, and the share of electric vehicles tripled.

1.1.2. The best solution for developing sustainable personal transportation.

It’s a common knowledge that petrol cars are one of the most harmful means of transport to the planet. As a result, several new technologies have been proposed to reduce CO₂ emissions, such as hybrid and electric cars. Yet electric bikes remain the most sustainable means of transport, which is reliable, and eco-friendly (Figure 1.1).

<i>Figure 1.1: Compare the carbon footprints of the different vehicles </i>

e-Bikes offer significantly extended travel distances compared to traditional bikes, thanks to the motor's power assistance. Research findings indicate that both walking and cycling (with bike riders and pedestrians) possess the potential to reduce car-related CO2 emissions by 8.5 million tonnes annually, equivalent to the energy consumption of 971,309 households over a year. In contrast, e-Bikes demonstrate a net emissions

</div><span class="text_page_counter">Trang 14</span><div class="page_container" data-page="14">

reduction capability of 16 million tonnes per year, nearly double the impact and surpassing the energy usage of 1.8 million homes for the same duration.

Additionally, e-bikes present another dimension of sustainability. They boast a lower carbon footprint than conventional bikes, as they require less food consumption to support pedalling. Traditional bikes have a carbon footprint of 21g of CO2 per kilometre, with a significant portion stemming from the additional food needed for cycling and the remainder originating from the manufacturing process [1]. The emissions from food production result from various factors, including deforestation for land use, livestock emissions, transportation, fertilizer production, and waste generation. 1.1.3. Definition of e-Bikes and e-MTB

An electric bicycle, commonly known as an e-Bikes, is a bicycle equipped with an electric motor that assists the rider in propulsion. There are 2 main types of electric bicycles classified according to the electric control signal:

• Pedal-Assist (No Throttle): Motor provides assistance only when the rider is pedalling.

• Throttle-Assist (No Pedalling Required): Motor can be activated by a throttle without pedalling and Pedalling is optional.

Before mentioning e-MTB, let talk about the mountain bikes (MTB). MTB are a bicycle designed for off-road cycling. Mountain bikes share some similarities with other bicycles, but incorporate features designed to enhance durability and performance in rough terrain, which makes them heavier, more complex, and less efficient on smooth surfaces. [2]

Therefore, e-MTB are a combination of the ruggedness of MTB and the electric motor of e-Bikes.

1.1.4. The development potential of e-MTB and the need for e-MTB conversion kits. Over the years, the sales of e-Bikes have increased rapidly due to the rise in fuel costs, which has led to the growth of the implementation of electric bicycles as a daily means of transport.

The electric MTBs market is a small part of the entire e-Bike market (5.77B vs 34.99B) (Figure 1.2 & 1.3). However, while the e-Bikes market already has many giants, especially Chinese companies, the e-MTB market still has opportunities for Vietnamese companies to develop. Besides that, although the market share is not large, the estimated Compounded Annual Growth Rate (CAGR) of e-MTB is 12.56% [3] , which is 1.5 times higher than that of e-Bikes (8.16%) [4]

</div><span class="text_page_counter">Trang 15</span><div class="page_container" data-page="15">

<i>Figure 1.2: Growth Trends & Forecasts </i>

The advantages of e-MTB include [5]: • Pedal assistance for longer distances • Optimal off-road capability

• Higher speeds, more fun • Everyday practicality

• Flexibility for different types of riders

However, there are disadvantages of e-MTB like [5]: • High cost

• Heavy weight

• Dependence on battery life

And the e-MTB conversion kit can efficiently deal with these three existing disadvantages of e-MTB on the market.

<b>1.2. Literature review </b>

1.2.1. Domestic research

At the current time, in May 2024, we have not found any research papers on domestic e-Bikes or e-MTB design. However, there are some domestic companies that have been developing their own lines related to e-MTBs, such as:

<b>• The Edge TX product line by Robot Viet Limited Liability Company: </b>

mid-drive conversions kit for MTB to e-MTB in prototype process.

<b>• The Stella product line by Robot Viet Limited Liability Company (or </b>

VierCycle): hub-drive conversions kit for bikes to e-bikes in commercial state.

</div><span class="text_page_counter">Trang 16</span><div class="page_container" data-page="16">

<b>• The MB1 product line by MET EV Company: e-Bikes in prototype process. </b>

1.2.2. International research

<i>The studies </i>

• In 2021, a group of students from the Maritime University of Raja Ali Haji in Indonesia designed and tested an electric-assisted bicycle with the aim of providing an alternative to traditional fossil fuel powered transportation for residents on a remote island [6]. The bicycle included components such as a 24 VDC 10Ah DC power source, a motor driver, a speed controller, a unidirectional freewheel in the front sprocket, a throttle, a 24 VDC 350 Watt brushed DC motor, and no torque sensor. The DC motor was mounted on the bicycle frame in an A-shaped arrangement, and the power source was placed on the saddle. By using a tachometer sensor and a voltmeter to measure wheel speed, motor voltage, and current consumption, the research group demonstrated a linear relationship between motor input voltage, motor current, and wheel speed. Real-world testing showed that users could ride the electric bicycle without pedalling, and the motor temperature increased from 30.6°C to 56.8°C within 21 minutes. However, the DC motor eventually burned out, and the research group attributed this to long-term motor usage and resulting overheating.

• A research article published in December 2022 summarized the characteristics and capabilities of Mid-drive e-Bikes and delved into the analysis of different motors for various electric bicycle applications [7]. The article highlighted the current trend of using permanent magnet synchronous motors (PMSM) powered by AC as the most suitable motor for e-Bikes due to their ability to generate high torque, especially at low speeds. The article also provided a Fuzzy control algorithm flowchart for controlling an e-Bikes in the presence of various disturbances such as uneven terrain and gusty winds.

<i>The commercial products </i>

• There are many famous and long-standing manufacturers of e-MTB products, the most typical are: BOSCH, YAMAHA, GIANT, TREK, SCOTT, ... Moreover, there are also several manufacturers of mid-drive conversion kit for bikes to e-Bikes from China, the most typical is BAFANG.

• There are also several commercial projects in the process of calling for capital on IndieGoGo since 2020 [8] related to (Figure 1.4)

</div><span class="text_page_counter">Trang 17</span><div class="page_container" data-page="17">

<i>Figure 1.4: A prototype mid-drive conversion kit to e-Bikes in IndieGoGo </i>

However, the conversion kit for MTB to e-MTB being commercialized on the market comes from only the company CYC-Motor in Taiwan (Researched in 2024

The price of an unibody e-MTB is not affordable with almost buyers. The average price of the best e-MTB of 2020 at €8,121 and increased to €8,846 for 2021. Moreover, survey participants spent an average of €4,593 on their current e-MTB while planning to spend €4,953. [10]

Without an official distributor and the market is not large enough, domestic maintenance, repair and replacement will certainly be very difficult.

</div><span class="text_page_counter">Trang 18</span><div class="page_container" data-page="18">

<b>1.4. Research and implementation range. </b>

This thesis presents the research and design of a conversion kit for MTB into MTB, with the following design criteria.

<b>e-Criteria 1: Working in Off-Road condition & Require power </b>

Assists cyclists on off-road terrain, with 10% gradients at speeds of 10-12 mph. The weight of the cyclist is 75kg and the weight of the bike is 20kg. The average power for the e-MTB effort is about 220 watts. (Table 1.1)

• Average speed = 10 -12 Mph and power of biker = 220 watts, which is equal to the average bike speeds [14] and average power consumption of cyclists using MTB without electricity [15]. This means that when the power assist mode is turned on, the pedaller’s speed and feeling can remain unchanged even though it is in tough biking conditions.

• Slope = 10% in off-road condition, which is arranged in difficult levels and only for mountain bikes. Not suitable for children under 11 years old. Requires a high level of competency in bike control and a high level of physical fitness. Technically challenging with features such as tight turns, small rock steps, narrow boardwalk sections, and may cross steep exposed side slopes. [16]

• The weight of the cyclist is about 75kg which is obtained from test methods section 6.1 TCVN 7448:2004 "Electric bicycles - General safety requirements and test methods".

• The weight of the bike is 75kg which is obtained from several previous prototypes.

<i>Table 1.1: Constraints for Criteria 1 in summary </i>

Biker’s power consumption ≤ 220𝑊

<b>Criteria 2: Pedal assistance </b>

The system has to support that the power is equal to the power consumed by the biker times the defined gain coefficient in real time

</div><span class="text_page_counter">Trang 19</span><div class="page_container" data-page="19">

<b>Criteria 3: Lower limit for maximum speed at crankset </b>

Motor speed at crankset is not slower than cadence of bikers. Avoid the situation where the biker pedals faster than the motor pulls.

<b>1.5. Structure of the thesis </b>

The thesis is organized as follows:

Chapter 1: Giving reasons why e-MTB is needed in green mobility and defining the design criteria.

Chapter 2: Choose main components/ parts/ modules to design and build a complete e-MTB.

Chapter 3: Calculation and mechanical design for transmission elements Chapter 4: Calculation and electronic design for several modules

Chapter 5: Design controller for the Field Oriented Control (FOC) method and build the concept for pedal assist.

Chapter 6: Experimental results and evaluation. Chapter 7: Summary and development orientations.

</div><span class="text_page_counter">Trang 20</span><div class="page_container" data-page="20">

<b>CHAPTER 2: SELECTION METHODS 2.1. Mapping selections </b>

<i>Figure 2.1: Priority selections </i>

<i>Figure 2.2: Secondary selections </i>

<b><small>ESC Open Source Rotor Position </small></b>

<b><small>Sensor Torque Sensor </small></b>

<b><small>Temperature Sensor </small></b>

<b><small>BMS & Cell PIN </small></b>

<b><small>Bluetooth Communication </small></b>

<b><small>Module </small></b>

</div><span class="text_page_counter">Trang 21</span><div class="page_container" data-page="21">

Following design criteria rooted in the interplay among system components, the

<i>selection process is bifurcated into two distinct parts: Priority Selections and Secondary </i>

<i>Selections. Priority Selections wield significant and direct influence over the system. </i>

<i>Conversely, Secondary Selections, while crucial, offer multiple alternatives or can be </i>

omitted without compromising the system integrity.

<b>2.2. Priority selections </b>

The selection process adheres to a hierarchical order, commencing with options characterized by minimal input arrows and progressing towards options influenced by a greater number of input arrows.

2.2.1. Position of actuator & working principal

Motors mounted on e-Bikes are categorized into two primary types: Hub-Drive and Mid-Drive. The figures below (Figure 2.3 & 2.4) illustrate these two layout options more explicitly, with the yellow circle representing the motor.

To elucidate why the majority of e-MTBs on the market adopt mid-drive configurations, the table below (Table 2.1) [17] will analyse the advantages and disadvantages of these two types of e-Bikes.

<i>Table 2.1: Comparison between Mid-drive and Hub-drive </i>

<b>MID - DRIVE </b>

<b>Power at the Crank: </b>

This central placement delivers power directly to the bike's chain and subsequently the rear wheel, which closely mimics the natural pedalling experience.

<b>Efficient Power Distribution: </b>

Provide more efficient power distribution, as they leverage the bicycle's gears.

<b>Better Handling and Balance: </b>

Typically offer improved handling and balance because the motor's weight is centralized, resulting in a more natural weight distribution.

<b>Ideal for Off-Roading: </b>

<b>Costlier and Complex: </b>

It is higher cost and complexity. Installation and maintenance might be more involved and costly compared to hub drive motors

<i>Figure 2.3: Mid-drive </i>

</div><span class="text_page_counter">Trang 22</span><div class="page_container" data-page="22">

It is often favoured for off-road and mountain biking, where power and control are essential. Their ability to adapt to different terrains and inclines makes them an excellent choice for adventurous riders.

<b>HUB - DRIVE </b>

<b>Simplistic Integration: </b>

It is integrated into the bicycle's wheel hub, either in the front or rear wheel. This design is simple and relatively straightforward, making hub motors an attractive choice for those who want a hassle-free e-Bikes experience.

<b>Low Maintenance: </b>

There are fewer moving parts compared to mid-drive motors. This can result in lower long-term ownership costs.

Suitable for Everyday Commuting:

Hub drive e-bikes are well-suited for urban and city commuting, as they provide consistent power delivery on flat terrain. They are also quieter than mid-drive motors, which can be an advantage in residential areas.

<b>Limited Efficiency on Inclines: </b>

Hub drive motors lack the gear-shifting capability of mid-drive motors, making them less efficient when tackling steep hills.

<b>Weight Distribution Concerns: </b>

Since the motor is located in the wheel hub, hub drive e-bikes can exhibit imbalanced weight distribution. This may affect handling and control, especially on rough terrain.

<i>With Criteria 1: Working in Off-Road condition, easily select mid-drive </i>

transmission for this conversion kit.

Drawing reference from various mid-drive principal diagrams available on the market, including patents, enables a comprehensive understanding of the design principles and operational mechanisms underlying mid-drive e-Bikes [18] (Figure 2.5).

<i>Figure 2.5: The way to measure torque of the right leg </i>

<i>1: Motor – 2 & 3: Clutch transmission device – 4: Axle – 5: Chainwheel – 6: Torque detection unit – 7: Hosing – 11 & 12: Gears – 41: Hollow tube – 42a & 42b: Cranks </i>

</div><span class="text_page_counter">Trang 23</span><div class="page_container" data-page="23">

The proposed principle diagram (Figure 2.6) must adhere to the following criteria: 1. The sensor position solely measures the power exerted by the cyclist and does

not gauge the total power applied to the vehicle.

2. The moments generated by the pedal and the motor, acting on the vehicle (rear wheel), remain independent of each other. Specifically:

• The diagram must prevent motor torque from being transmitted to the pedal to ensure safety when only the motor is in operation.

• Prevents pedal torque from acting on the motor, thereby ensuring that the pedaller does not expend additional effort to rotate the motor when electric assistance is not engaged.

<i>Figure 2.6: Proposed principle diagram </i>

<i>The system will have 2 operating states: Without power assistance (Figure 2.7) and </i>

<i>with power assistance (Figure 2.8). The image below (Figure 2.9 &2.10) will more </i>

</div><span class="text_page_counter">Trang 24</span><div class="page_container" data-page="24">

clearly describe the torque line (Red line) and how the sensor measures the runner's torque.

2.2.2. Battery cover selection

Due to running in relatively harsh hilly terrain, priority is given to choosing an aluminium protective cover for the battery, with the aim of reducing weight, providing better protection and finally good heat dissipation. Like other electric vehicle manufacturers, aluminium is widely used in the construction of battery boxes.

There are aluminium protective covers for electric vehicles available on the market and have a bottle design [19]. The capacity of the battery depends on the cells we choose to package. However, the general configuration of the battery is 10s2p or 10s3p and the rated voltage is 36V.

2.2.3. Motor selection 2.2.3.1. Select motor type

Modern electric bicycles (e-Bikes) predominantly utilize brushless DC (BLDC) motors, a significant advancement over the older brushed motor technology. The primary distinction between these two types of motors lies in the method of current direction alternation. In brushed motors, mechanical brushes are used to switch the direction of the current flowing to the motor, a process that inherently reduces efficiency

</div><span class="text_page_counter">Trang 25</span><div class="page_container" data-page="25">

and leads to wear and tear over time. In contrast, BLDC motors achieve this alternation electronically, eliminating the need for brushes and thereby enhancing efficiency and longevity. This technological shift has led to the widespread adoption of BLDC motors in e-Bikes over the past decade, solidifying their status as the industry standard due to their superior performance and durability [20]. In addition, their electronic commutation not only enhances efficiency and durability but also allows for more precise control compared to traditional brushed motors. Consequently, BLDC motors have become the dominant choice in the design and development of contemporary electric bicycles. 2.2.3.2. Motor selection & specifications

<i>Calculate power to satisfy Criteria 1 by OMNI Calculator [21] the calculation </i>

formula is based on analysis and data collection of millions of people around the world. The results have shown that a total of 𝑃_{𝑑𝑒𝑠𝑖𝑟𝑒} ≈ 727𝑊 is needed, where power of biker: 𝑃_{𝑏𝑖𝑘𝑒𝑟} ≤ 220𝑊 then power of motor: 𝑃_{𝑚𝑜𝑡𝑜𝑟} ≥ 𝑃_{𝑑𝑒𝑠𝑖𝑟𝑒}− 𝑃_{𝑏𝑖𝑘𝑒𝑟} ≥ 507𝑊. The detail result in section 3.2.1. Calculate the power for the motor and motor selection.

To meet the proposed capacity requirements, the operating voltage is set at 36V. However, considering factors such as domestic availability and the need for a compact

<b>design, the Genuine Ryobi RY18LMX40A Electric Motor emerges as a suitable </b>

choice. This motor, with a rated capacity of approximately 600W, not only aligns with the voltage specification but also offers a reliable and efficient solution that is readily accessible in the domestic market. Its compact design further enhances its suitability for the intended application, ensuring both performance and practicality.

<i>Figure 2.9: Product drawing from the RYOBI manufacturer </i>

Φ Φ Φ Φ

</div><span class="text_page_counter">Trang 26</span><div class="page_container" data-page="26">

From the manufacturer's motor specifications (Detail specification in section 3.2.1. Calculate the power for the motor and motor selection, max speed at rated torque at the rated voltage 36V:

Max Speed @ Rated Torque (Rpm) = 3230 rpm 2.2.4. Control method selection

2.2.4.1. Compare pros and cons of control methods

Nowadays, there are many control strategies for PMSM drives, including oriented control (FOC), direct flux control (DFC), and direct torque control (DTC) [22]

field-To choose the most suitable control type, we will look at the following set of three main factors: Adaptation with load variation, EV applications, large user community.

<i>Table 2.2: Comparison between current control methods </i>

<b>ADAPTATION WITH LOAD </b>

very large just

research ^{some practical}projects

2.2.4.2. Working principle of FOC

The main idea of vector control is to control not only the magnitude and frequency of the supply voltage but also the phase. In other words, the magnitude and angle of the space vector are controlled [24]. The steps to implement FOC can be divided into four main steps:

<i>Step1: Measure current already flowing in the motor </i>

Just measure two phases and can interpolate the remaining phase using Kirchhoff's current law. (Figure 2.10)

</div><span class="text_page_counter">Trang 27</span><div class="page_container" data-page="27">

<i><b>Figure 2.10: How to measure the current of the phases in FOC </b></i>

<i>Step2: Compare the measured current (vector) with the desired current (vector), and generate error signals </i>

Since a synchronous motor, if we know the angle of the rotor, we can know the angle of the rotor flux. The error signal is indicated in the figure below (Figure 2.11).

<i><b>Figure 2.11: Error between measured current vector and desired current vector </b></i>

</div><span class="text_page_counter">Trang 28</span><div class="page_container" data-page="28">

Put i_s & Rotor flux axis into new coordinate systems (Figure 2.12 & 2.13). The purpose of this is to convert from sine form to linear form, it is a lot easier to regulate.

<i>Figure 2.12: Converting to the fixed coordinate </i>

<i>Figure 2.13: Converting to the synchronous rotation coordinate </i>

The above is an explanation in terms of graphics (Figure 2.14), but in terms of mathematics, we have the operation to convert the coordinate system:

<i>Figure 2.14: States of converting 3-phase to controllable 2-phase in FOC </i>

Forward Clark transformation, converse 3-phase to 2-phase 𝑖_𝛼 =³

2^𝑖^𝑎𝑖_𝛽 =^√3

2 ^𝑖^𝑏 ⁻√3

</div><span class="text_page_counter">Trang 29</span><div class="page_container" data-page="29">

To maximize torque and efficiency, the vector 𝑖_𝑑 has to become zero. It causes the vector 𝑖_𝑑 will lie on the q axis which is orthogonal with the rotor flux axis (Figure 2.15).

<i>Step 4: Modulate the correction voltages onto the motor terminals (Figure 2.16) </i>

<i>Figure 2.16: States of converting controlled 2-phase to 3-phase in FOC </i>

Reverse Park transformation:

𝑣_𝛼 = 𝑣_𝑑𝑐𝑜𝑠𝜃_𝑑− 𝑣_𝑞𝑠𝑖𝑛𝜃_𝑑𝑣_𝛽 = 𝑣_𝑑𝑠𝑖𝑛𝜃_𝑑+ 𝑣_𝑞𝑐𝑜𝑠𝜃_𝑑Reverse Clark transformation:

𝑣_𝑎 =²3^𝑣^𝛼𝑣_𝑏 = −¹

3^𝑣^𝛼 ⁺1√3^𝑣^𝛽

</div><span class="text_page_counter">Trang 30</span><div class="page_container" data-page="30">

𝑣_𝑐 = −¹3^𝑣^𝛼⁻

In addition, in FOC there are also many other control techniques, that can operate such as observer, tracking and torque optimization (Figure 2.17).

<i>Figure 2.17: Additional reluctance torque of Toyota/ Prius Hybrid THS II Motor [25] </i>

FOC can also operate in sensor-less mode, which means that without a rotor

<i>position sensor, an observer can still predict rotor position for Step 2, yet this is not </i>

recommended for high load cases or the initial state of the motor. This diagram will illustrate (Figure 2.18) the way for estimating rotor position 𝜃̂.

<i>Figure 2.18: Stationary frame state observer for a salient machine [26] (LPF: Low Pass Filter) </i>

</div><span class="text_page_counter">Trang 31</span><div class="page_container" data-page="31">

2.2.5. Gearbox selection 2.2.5.1. Select gearbox type

<i>Criterion 3 combined with experimental data and collected from sources on the </i>

Internet [27], the maximum pedalling cadence of a cyclist: n_cadence ≤ 130 rpm. The transmission ratio of the entire system can be calculated using the formula:

𝑢_{𝑠𝑦𝑠𝑡𝑒𝑚} ≤ ^𝑛^{𝑚𝑜𝑡𝑜𝑟}𝑛_{𝑐𝑎𝑑𝑒𝑛𝑐𝑒} ⁼

2.2.5.2. Clutch in gearbox selection

As mentioned in the principal diagram, in section 2.2.1. Position of actuator & working principal, there needs to be a clutch inside the gearbox to prevent the cyclist from pulling the motor when running in unassisted mode.

<i>Table 2.3: Comparison between types of clutches in gearbox </i>

<b>W One-way bearing Freewheel Sprag Clutch LOAD </b>

<b>CAPACITY </b> ^40%

slip at heavy load broken at heavy

<small>1 There are commercially available sprockets on the market. With large chain sprockets according to BCD 104 standards (For MTB), the tooth number ranges from 32 to 46 teeth. With small sprockets specialized for e-Bikes with 9/13 teeth. From there, the external transmission ratio can be calculated from 32/13 to 46/9. </small>

</div><span class="text_page_counter">Trang 32</span><div class="page_container" data-page="32">

<b>SCORE 0.825 0.35 0.85 </b>

Sprag Clutch is an optimal choice. Furthermore, some types of Sprag Clutch combined with planetary gears are also available on the market (Max torque is 30Nm).

<i>Figure 2.19: Product drawing from the manufacturer </i>

2.2.6. Crankset system selection 2.2.6.1. Select drive transmission type

For use in applications requiring very large loads and high transmission ratios, be sure to choose a chain drive instead of a belt drive.

2.2.6.2. Clutch in crankset selection

This clutch has the function of ensuring safety when the engine is operating and must have it so that the torque of the engine and the driver can be separated, from which the torque/ power of the operator can be measured cyclists.

Unlike the clutch in a gearbox, the priority criteria for selecting this clutch will be different. For example, the load capacity will not have a large weight because the torque that this clutch must load is only the pedal's torque, it does not include the motor's power.

<i>Table 2.4: Comparison between types of clutches in crankset </i>

<b>W ^One-way_bearingFreewheel Sprag Clutch DIFFICULTY </b>

a lot of choices a lot of choices no diversity

<b>VIBRATION 20% </b> _{very low}¹ _high^0.5 _{very low}¹<b>LOAD </b>

<b>CAPACITY </b> ^10%

slip at heavy load broken at heavy

</div><span class="text_page_counter">Trang 33</span><div class="page_container" data-page="33">

<b>SCORE 0.875 0.85 0.75 </b>

It can be said that using a one-way bearing as a clutch is the most suitable, although there are difficulties in design because the bearing size is often large and can cause entanglement in the pedal probe.

<b>2.3. Secondary selection </b>

2.3.1. Torque sensor selection

The first wave of e-bikes relied on simple cadence-based control resulting in quite unpredictable handling when cornering and during start/stop. Complementing this with a torque sensor to sense the force put into the pedals is a significant improvement.

On an e-bike, the torque sensor measures the applied pedalling torque of the rider and sends it to the controller of the e-bike along with cadence information. Based on this information an algorithm controls the motor speed and power. The performance of the e-bike as perceived by the rider depends heavily on the precision and accuracy of the motor control.

<i>Table 2.5: Comparison between types of torque sensors </i>

<b>W _{Strain Gauge}^Contactless^{Slipring Strain}_Gauge^{Indirect Strain}_GaugeSHAFT </b>

</div><span class="text_page_counter">Trang 34</span><div class="page_container" data-page="34">

<b>OTHER </b>

only hub-drive; no belt drive; limited dynamic

range

As evident from the overview table, the contactless strain gauge sensor is a quite favourable solution for many e-bike applications. The main disadvantage is that the rotating part needs additional electronics.

<b>The T13 Torque Sensor from Shanghai Moreway International Trade Company Limited was chosen because it is a contactless strain gauge. In addition, this </b>

sensor has a cadence signal (one phase in clockwise) to calculate the pedal's power, available stock and reasonable price.

<i>Table 2.6: Specification of T13 torque sensor </i>

2.3.2. Temperature sensor selection

The only criteria for choosing a temperature sensor is that it must be very small because it needs to be placed in the PCB circuit and in the motor stator. Therefore, NTC 10k resistant sensor is chosen.

2.3.3. Rotor position sensor selection

<i>Table 2.7: Comparison between types of rotation position sensors </i>

1 Bottom Bracket (BB) length 83 𝑚𝑚

4 Max. power consumption < 0.5𝑊

5 Speed signal ^{Forward rotation: 18 pulse/ rotation}

Backward rotation: no speed signal output 6 Torque output range 1.5 ~ 3.0 𝑉𝐷𝐶

</div><span class="text_page_counter">Trang 35</span><div class="page_container" data-page="35">

<b>HALL SENSOR ABI SENSOR ^SINUSOIDAL_SENSOR</b>

<b>ADVANTAGES </b>

- Simple and small construction

- Working in harsh environments - Low cost

<b>- Relatively high </b>

resolution <b>^{- High resolution}</b>and accuracy compared to incremental

encoders

continuous

feedback without discrete pulses

<b>DISADVANTAGES </b>

<b>- Low resolution </b>

compared to other type encoders - Susceptible to interference from external magnetic - Limited accuracy and precision, especially at high speed because of a small wrong in any phases

<b>- High cost </b>

- Requires more space for sensor placement

<b>- High cost </b>

- Requires more sophisticated signal processing - Requires accurate design for sensor mounting location

For durable, economic and streamlined design, I choose an option using hall sensor for Rotation position sensor

Problem solving plans:

• Just using the hall sensor for motors at low speed to reduce error caused by phase difference

• Using capacitors for filtering noise Hall sensor selection criteria:

• Type unipolar switching output instead of analog voltage output • High-temperature operation ≥ 80°C

• Ease to buy and low cost

• Output rise & fall time ≪ phases circle time 2.3.4. Select ESC open source

Several well-known brands provide Field-Oriented Control (FOC) controllers for various applications, including motor control, inverters, and other electronic systems. Some common examples include:

• Texas Instruments: Texas Instruments offers FOC controllers for electric motor applications, including both direct and indirect control.

</div><span class="text_page_counter">Trang 36</span><div class="page_container" data-page="36">

• Infineon Technologies: Infineon provides FOC controllers for a wide range of applications from electric motors to inverters.

• STMicroelectronics: STMicroelectronics offers FOC control solutions for electric motors and various other applications.

• NXP Semiconductors: NXP provides FOC controllers for motor and inverter applications.

• Microchip Technology: Microchip offers FOC control solutions for diverse applications ranging from electric motors to low-power electronics.

Additionally, there are many other companies such as Bosch, Mitsubishi Electric, and Toshiba that also provide FOC controllers for the market. However, to easily modify the program to suit the e-MTB application, two common open-source FOC controllers are considered: VESC and SimpleFOC.

<i>Table 2.8: Comparison between two popular FOC open sources </i>

<b>EASE TO PROGRAM 25% </b> using ChibiOS for ⁰ ¹

running multi-thread ^{framework Arduino}

<b>DEEP EMBEDDED </b> 25%

all firmware is public,

user can modify ^{a library on framework}Arduino cant modify

2.3.5. Select BMS & Cell pins

To achieve the highest capacity, INR18650-35E (3500mAh) cells from Samsung or INR18650-35V (3500mAh) cells from EVE are used. Both types of cells offer a maximum discharge current of approximately 10A. While the price of each EVE cell is two-thirds that of a Samsung cell, their quality is generally comparable. The battery configuration consists of two blocks connected in parallel, resulting in a peak current output of 20A. Consequently, a Battery Management System (BMS) rated at 10s 20A is selected to accommodate this configuration and ensure optimal performance and safety.

</div><span class="text_page_counter">Trang 37</span><div class="page_container" data-page="37">

<b>CHAPTER 3: MECHANICAL DESIGN </b>

The figure below (Figure 3.1) shows the mechanical principal diagram of the entire system, which was shown in section <i>2.2.1. Position of actuator selection & working </i>

chain-drive, crankset, and motor holder. Each section will detail the specific design considerations and methodologies used to optimize the performance and integration of these components within the overall system.

<i>Figure 3.1: Working principle & main blocks for mechanical design </i>

</div><span class="text_page_counter">Trang 38</span><div class="page_container" data-page="38">

<b>3.1. Working principle </b>

3.1.1. Method to measure torque from both legs

The operating principle of the bottom bracket in this design is to measure the total torque generated by both legs of the rider. This is achieved through the integration of torque sensors within the bottom bracket assembly.

<i>Figure 3.2: How the system measures the </i>

3.1.2. Method motor assist the system

The motor in this e-MTB design can operate in two scenarios: either faster than the pedal foot or at the same speed as the pedal foot. Due to Criterion 3 , there is no scenario where the motor operates slower than the pedal foot. The crankset clutch plays a crucial role in ensuring the safety and functionality of the system. Its primary functions are twofold:

• Protection of Pedal Legs: The crankset clutch safeguards the rider's legs from the impact of the motor. By engaging or disengaging as needed, it prevents the motor from transmitting excessive force to the pedals, thereby reducing the risk of injury or discomfort to the rider (Figure 3.4).

• Accurate Torque Measurement: The crankset clutch also ensures that the torque sensors only measure the rider's pedaling force, not the motor's torque. When the clutch is disengaged, it decouples the motor's torque from the crankset, allowing the sensors to always measure only the rider's input (Figure 3.5).

</div><span class="text_page_counter">Trang 39</span><div class="page_container" data-page="39">

<i>Figure 3.4: How does the system separate the torque of the motor and the </i>

<i>cyclist </i>

<i>Figure 3.5: How does the system assist the cyclist </i>

<b>3.2. Motor selection and transmission ratio </b>

3.2.1. Calculate the power for the motor and motor selection The system efficiency to crankset, not to rear wheel

𝜂 = 𝜂_{𝑔𝑒𝑎𝑟}<small>2</small> × 𝜂_{𝑐𝑙𝑢𝑡𝑐ℎ}× 𝜂_{𝑐ℎ𝑎𝑖𝑛} × 𝜂_{𝑏𝑒𝑎𝑟𝑖𝑛𝑔}³𝜂 = 0.97<small>2</small>× 0.99 × 0.93 × 0.99<small>3</small> = 0.84 Where: 𝜂_{𝑔𝑒𝑎𝑟} = 0.97: gear efficiency 𝜂_{𝑐𝑙𝑢𝑡𝑐ℎ} = 0.99: clutch efficiency 𝜂_{𝑐ℎ𝑎𝑖𝑛} = 0.93: chain efficiency

𝜂_{𝑏𝑒𝑎𝑖𝑟𝑛𝑔} = 0.99: 𝑎 pair of bearing efficiency

The required power to satisfy Criterion 1 is calculated using the formula form Omni Calculator [21]

• Resisting force due to gravity

𝐹_𝑔 = 𝑔 × sin(arctan(𝑠𝑙𝑜𝑝𝑒)) × (𝑀 + 𝑚) 𝐹_𝑔 = 9.807 × sin (arctan(10%) × (75 + 20) 𝐹_𝑔 = 92.701𝑁

Where: 𝑔 = 9.807 𝑚/𝑠<small>2</small>: gravitational acceleration 𝑠𝑙𝑜𝑝𝑒 = 10%: slop of hill

𝑀 = 75 𝑘𝑔: biker<small>′</small>s weight

</div><span class="text_page_counter">Trang 40</span><div class="page_container" data-page="40">

𝑚 = 20 𝑘𝑔: bike^′s weight • Rolling resistance

𝐹_𝑟 = 𝑔 × cos(arctan(𝑠𝑙𝑜𝑝𝑒)) × (𝑀 + 𝑛) × 𝐶_𝑟𝑟

𝐹_𝑟 = 9.807 × cos(arctan(10%)) × (75 + 20) × 0.0253 𝐹_𝑟 = 23.453𝑁

Where: 𝑔 = 9.807 𝑚/𝑠²: gravitational acceleration 𝑠𝑙𝑜𝑝𝑒 = 10%: slop of hill

𝐶_𝑟𝑟 = 0.0253: coefficient at knobby tires & offroad surface • The aerodynamic drag

𝑃 =(𝐹_𝑔+ 𝐹_𝑟 + 𝐹_𝑎) × 𝑣1 − 𝑙𝑜𝑠𝑠

𝑃 =(92.701 + 23.453 × 7.714) × 5.5561 − 4%

𝑃 = 716.886𝑊

Where: 𝐹_𝑔 = 92.701𝑁: Resisting force due to gravity 𝐹_𝑟 = 23.453𝑁: Rolling resistance force 𝐹_𝑎 = 7.714𝑁: Aerodynamic drag

𝑣 = 5.556 𝑚/𝑠: biking speed

𝑙𝑜𝑠𝑠 = 4%: percentage loss in power at chain not well − oil. The require power for motor:

𝑃_{𝑚𝑜𝑡𝑜𝑟} =^{𝑃 − 𝑃}^{𝑐𝑦𝑐𝑙𝑖𝑠𝑡}𝜂

P_motor ≥^{716.886 − 220}

Where: 𝑃 = 716.886𝑊: Power to satisfy design criteria 𝑃_{𝑐𝑦𝑐𝑙𝑖𝑠𝑡} ≤ 220𝑊: Power the cyclist wants to expend 𝜂 = 0.84: the system efficiency

As the previous preliminary calculation in section 2.2.3. Motor selections, the

<b>engine selected is Genuine Ryobi RY18LMX40A Electric Motor with specification </b>

table (Table 3.1).

</div>