Optimization of the Intake System for the 2008 Formula SAE Race Car

Transcription

1 Optimization of the Intake System for the 2008 Formula SAE Race Car Nilufar Damji A thesis submitted in partial fulfillment of the requirements for the degree of BACHELOR OF APPLIED SCIENCE Supervisor: Professor M. Bussmann Department of Mechanical and Industrial Engineering University of Toronto March 2008

2 Abstract The purpose of this thesis project is to develop an optimal intake system design for the University of Toronto (U of T) Formula SAE racecar. Computer simulations using Ricardo Wave and a vehicle simulation program, as well as physical tests using an engine dynamometer (dyno), were conducted to develop a better understanding of the intake system parameters on the output performance of the engine. The focus of this research project was to determine the effects of different intake runner lengths and different plenum volumes on the torque and power outputs of the Honda CBR600 F4i motor. Design objectives, design constraints, and benchmarking of previous systems were combined to determine the plenum style and geometrical specifications to use to develop the design of the intake system. The optimization of the intake system design for the 2008 U of T Formula SAE racecar was determined to be achieved if the output performance of the 2008 intake system design exceeded that of the 2007 engine system in terms of torque and power. The results of the computer simulations, for Ricardo Wave and the vehicle simulation program, were unable to make a significant distinction between different intake system configurations. Consequently, the simulation data was not useful in evaluating system performance, and therefore unable to aid in system optimization. As a result, a great deal of physical testing on the dyno was used to determine the optimal intake system configuration. These results were also used to determine the effects of various intake system parameters on the torque and power outputs of the motor. A cone style intake system was designed for use on the 2008 racecar. However, from analysis of the dyno results, it was determined that this system did not perform as well as the 2007 engine system. Several modifications were made to the original cone style intake system design in an effort to improve its performance. However, this system was also unable to perform as well as the 2007 engine system. Therefore, it was decided to use the 2007 intake and exhaust systems for the 2008 racecar. In order to achieve better output performance than the 2007 engine system, a California specification intake camshaft was installed in the motor. This increased the overall torque and power values of the 2007 engine system throughout the drivable engine speed range. Therefore, although the expected outcome was not achieved, the goals of the thesis project were ultimately attained. 1

3 Acknowledgements I would like to take this opportunity to acknowledge and thank the individuals that provided help and support throughout the completion of this thesis: Professor Bussmann for being my thesis supervisor and for providing excellent support and feedback for this project Axis Rapid Prototyping for sponsoring the Formula SAE team with the manufacture of the intake and fuel systems Profile Waterjet for sponsoring the Formula SAE team with the manufacture of the throttle body plates Kenspin for sponsoring the Formula SAE team with the manufacture of the bellmouths Neal Persaud for his endless support and guidance throughout the four years I have been a member of the FSAE team James Correia, Nick Burgwin, and Saqib Siddiqi for their help with operating, trouble shooting, and maintaining the dynamometer Leonardis Simonis for his help designing the fuel system and developing the relationship with Axis Rapid Prototyping Mark Osmokrovic for his help with packaging of the intake system in the chassis Vince Libertucci for his development of the vehicle simulation program Andrew Wong for his excellent photography skills To all past and present FSAE members that have helped me become the engineer I am today. Thank you for giving me the opportunity to develop my skills and thank you for supporting me throughout my years on the team i

4 Table of Contents Acknowledgements... i List of Symbols... iv List of Figures... v List of Tables... vi 1 Introduction Purpose of Thesis Project Formula SAE The Formula SAE Competition Formula SAE Rules Affecting the Intake System Design Motivation for the Project Background Information Design of the 2008 University of Toronto Formula SAE Race Car The Intake System Literature Review Objectives Design Objectives Design Constraints Design Methodology Benchmarking of Previous Systems Ricardo Wave Dynamometer Manufacturing Methods Initial Intake System Design Plenum Style Modular Intake System Design for Dyno Testing Testing Ricardo Wave Simulation Model Dynamometer Testing Vehicle Simulation Results and Analysis Results of the Ricardo Wave Simulation Results of the Dynamometer Testing Important Notes to Consider When Interpreting the Dyno Results Results of Runner Length Testing Results of Plenum Volume Testing Results of Camshaft Testing Results of Runner Length Testing with the Cali Spec Intake Camshaft Results of Primary Length Testing with the Cali Spec Intake Camshaft Comparison of the Modular Intake System and the 2007 Engine System Design of the 2008 Intake System Vehicle Simulation Results Validation of Optimal Design Conclusion References ii

5 Appendix A Formula SAE Rules for Powertrain Design [1] Appendix B Material Properties of Duraform Polyamide [4] iii

6 List of Symbols C degrees Celsius CBR600 F4i cc ECU FEA FSAE ft ft-lbs hp Hz in (or ) K kpa km L mm psia rpm SAE sec SLS UTXXXX U of T USD model of Honda motorcycle used for the U of T FSAE racecars cubic centimeters engine control unit finite element method Formula Society of Automotive Engineers foot foot-pound horsepowe hertz inches Kelvin kilopascals kilometers liter millimeter pounds per square inch atmospheric, unit of pressure revolutions per minute Society of Automotive Engineers seconds selective laser sintering rapid prototyping method U of T FSAE car with XXXX as the year designation University of Toronto Currency used in the United States of America iv

7 List of Figures Figure 1: 2007 University of Toronto Formula SAE Racecar.. 1 Figure 2: Envelope Area Constraining the Intake System Placement.. 3 Figure 3: Water Brake Dyno Used for Physical Testing.. 6 Figure 4: Components of the Intake System. 7 Figure 5: 2004 Racecar Showing the Trapezoidal Intake System 8 Figure 6: Log Style Intake System 8 Figure 7(a): FEA of the Log Style Intake for Displacement in the Axial Direction. 17 Figure 7(b): FEA of the Log Style Intake for Displacement in the Radial Direction Figure 8: FEA of the Cone Style Intake for Displacement Figure 9: Three Dimensional Model of the Modular Intake System 18 Figure 10: Modular Intake System Being Tested on the Engine Dyno 19 Figure 11: Screen Shot of the UT2007 Intake System Model in Ricardo Wave.. 20 Figure 12: Bent Intake Runner Sections Required for Ricardo Wave Model Figure 13: Comparison of Torque Curves for the 2007 Engine System Figure 14: Torque Curve Produced in Ricardo Wave for the 2007 Engine System. 25 Figure 15: Torque Curves for Intake Runner Length Testing Figure 16: Power Curves for Intake Runner Length Testing 29 Figure 17: Torque Curves for Plenum Volume Testing Figure 18: Power Curves for Plenum Volume Testing. 31 Figure 19: Camshaft Profile Comparison of the Cali Spec and Stock Intake Camshafts. 32 Figure 20: Torque Curves for Camshaft Testing.. 33 Figure 21: Power Curves for Camshaft Testing 34 Figure 22: Torque Curves for Intake Runner Length Testing with the Cali Spec Intake Camshafts. 35 Figure 23: Power Curves for Intake Runner Length Testing with the Cali Spec Intake Camshafts. 36 Figure 24: Torque Curves for Exhaust Primary Length Testing with the Cali Spec Intake Camshafts. 38 Figure 25: Power Curves for Exhaust Primary Length Testing with the Cali Spec Intake Camshafts. 38 Figure 26: Comparison of Torque Curves for the Modular Intake System and the 2007 Engine System. 40 Figure 27: Comparison of Power Curves for the Modular Intake System and the 2007 Engine System. 40 Figure 28: Final Intake System Design. 43 Figure 29: Packaging of the Intake System Within the Envelope of the 2008 Chassis 43 Figure 30: 2008 Intake System Being Tested on the Engine Dyno.. 44 Figure 31: Comparison of Torque Curves for the 2008 Intake System and the 2007 Engine System. 45 Figure 32: Comparison of Power Curves for the 2008 Intake System and the 2007 Engine System. 45 Figure 33: Packaging of the 2008 Optimized Intake System within the Envelope of the 2008 Chassis 53 v

8 List of Tables Table 1: Point Breakdown for Each Event Table 2: Advantages and Disadvantages of the Three Intake System Designs Table 3: Intake and Exhaust Characteristics of Previous Systems Table 4: Summary of Previous Drivers Experiences With Past Intake Systems. 13 Table 5: Intake System Parameters Tested on the Dyno Table 6: Summary of Torque and Power Output Results for Intake Runner Length Testing 29 Table 7: Summary of Torque and Power Output Results for Plenum Volume Testing 31 Table 8: Summary of Torque and Power Output Results for Camshaft Testing.. 34 Table 9: Summary of Torque and Power Output Results for Intake Runner Length Testing with the Cali Spec Intake Camshaft Table 10: Summary of Torque and Power Output Results for Exhaust Primary Length Testing with the Cali Spec Intake Camshaft Table 11: Summary of Torque and Power Output Results for the Comparison of the Modular Intake System and the 2007 Engine Systems Table 12: Summary of Torque and Power Output Results for the Comparison of the 2008 Intake System and the 2007 Engine Systems Table 13: Corner Lap Times for Each Torque Curve Table 14: Straight and Total Lap Times for Each Torque Curve. 49 Table 15: Weight Comparison of the Cali Spec and Stock Intake Camshafts.. 51 vi

9 1 Introduction 1.1 Purpose of Thesis Project The purpose of this thesis project is to develop an optimized intake system design for the 2008 University of Toronto Formula SAE racecar (UT2008). The objectives of the research results are to develop a better understanding of the effects of various intake system parameters on the output performance on the racecar. This will be accomplished through a combination of computer simulations and physical testing of several different intake system designs. 1.2 Formula SAE The Formula SAE Competition Formula SAE (FSAE) is a design competition in which teams of university students design, manufacture, and race, small open-wheeled, formula-style racecars. UT2007, which can be seen in Figure 1, is an example of the type of vehicles that are used in this Formula series. Figure 1: 2007 University of Toronto Formula SAE Racecar The purpose of the competition is to develop a prototype racecar for a fictional manufacturing firm that will be evaluated for use as a production item. The total cost of the car must not exceed $25,000 USD [1]. There is also a manufacturing requirement for 1

10 limited production quantities of four vehicles per day [1]. The car is to be designed for a nonprofessional weekend autocross racer. The vehicles are evaluated in terms of their ontrack performance, their feasibility for manufacturing, and their marketability for the specified audience. The cars are judged in both static and dynamic events where a maximum score of 1000 points can be achieved. The point breakdown for each even can be seen in Table 1. The team that achieves the highest score is the winner of the competition. Table 1: Point Breakdown for Each Event [1] Static Events Maximum Points Available Presentation 75 Engineering Design 150 Cost Analysis 100 Dynamic Events 75 Dynamic Events Acceleration 75 Skid-Pad 50 Autocross 150 Fuel Economy 50 Endurance 350 Total: 1000 Points A majority of the points can be obtained through exceptional performance in the dynamic events, especially in the endurance event. The vehicles must be reliable and have excellent dynamic capabilities to achieve a high score in this event. Top speed in a straight line is only one aspect of the competition, which is judged in the acceleration event. However, the autocross and endurance events, which account for 50% of the maximum achievable points, require a combination of optimized powertrain and suspension system designs. The tracks for these events include many obstacles, such as slaloms, large and tight radius corners, and hairpins, separated by straight sections. This is a critical design consideration for determining the optimal torque curve for the motor. Instead of aiming to have the peak power output at the maximum engine speed, it is more important to produce sufficient torque throughout the engine operating range to allow the car to transition easily from a corner to a straight section. 2

11 1.2.2 Formula SAE Rules Affecting the Intake System Design SAE has formulated an extensive list of rules that govern the competition, which are mostly for safety reasons due to the nature of the competition. The complete list of rules that influence the powertrain design can be found in Appendix A. The rules that directly affect the intake system design include the following [1]: The engine must be a four stroke piston engine having a displacement of no more than 610cc The engine must breathe through a 20mm restrictor (if using 94 octane fuel) All parts of the engine air system must lie within the envelope the surface defined by the top of the roll bar and the outside edge of the four tires. Figure 2 shows the area that is described by the envelope. These rules must be taken into account when designing the many components of the intake system. Figure 2: Envelope Area Constraining the Intake System Placement [1] 1.3 Motivation for the Project The purpose of the FSAE competition from the perspective of the students is to gain as much valuable learning experience as possible during their university years. Most of the competition judges are prominent people in industry that are very knowledgeable about specific aspects of racecar design. During the design event, these judges ask numerous 3

12 questions to the students in order to determine the extent to which the team members understand the design of their racecars. Therefore, although a huge emphasis is placed on how well a car is built and how well it performs on track, there is also much consideration given to the engineering design of the vehicle. Each member of the U of T FSAE Team must understand the fundamental principles of every design. This is important to ensure the continuity of knowledge and success in future seasons. It is also critical that the team is able to justify their designs through computer simulations or physical tests. Therefore, one of the goals of this thesis project is to determine how changes to the intake runner lengths and plenum volume affect the torque and power outputs of the motor. Since limited research has been conducted on intake system designs for FSAE vehicles, the data that will be gathered for this project will have a great impact on future designs. 4

13 2 Background Information 2.1 Design of the 2008 University of Toronto Formula SAE Race Car UT2008 was based off the Honda CBR600 F4i engine. This motor was chosen because it has been used for several years on the U of T racecars. Consequently, a significant amount of knowledge about this particular motor has been developed. It is hoped that the results of this project will lead to a better understanding of the intake system design parameters on the output of the motor. There are several engine system design initiatives for the 2008 season. For the first time, an engine simulation software program, called Ricardo Wave, will be used to help improve the efficiency of conducting many design iterations. The advantage of using software to simulate the operation of the engine is that multiple designs can be tested without having to fabricate and physically test each one. This reduces both the time and costs associated with the preliminary design process. In order to ensure that the simulation model is correct, some of the results of the simulation must be compared to the results of a physical test. However, since combustion is a complex event that cannot be repeated to a fine precision, and certain geometries are difficult to model, the absolute values of the results from the simulation and the physical test may not necessarily correlate. However, the profile of the torque and power curves, as well as the trends determined by changing various parameters, will be the same for the simulation and the physical test. The physical testing of the intake system designs will be conducted using an engine dyno. A water brake dyno, which can be seen in Figure 3 (on page 6), will be used to conduct this testing. This type of dyno uses an impeller to generate a brake torque that holds the engine at a user specified speed and load. A load cell located on the dyno arm is used to calculate the value of the brake torque. This value is recorded using the engine control 5

14 unit (ECU) and is plotted on a graph against the instantaneous speed of the motor, thereby producing a torque curve. Figure 3: Water Brake Dyno used for Physical Testing [2] In order to determine the optimal intake system design in terms of vehicle performance, the torque curves produced on the dyno will be used in a custom made vehicle simulation program developed by the U of T FSAE suspension team. This program calculates the total lap time of the vehicle, as well as the lap times in various radius corners and different length straight sections. The number of shifts required per lap can also be determined. This program will therefore be used to measure the performance and drivability of various intake system designs, which will help determine the optimal intake system. 2.2 The Intake System The intake system is one of the most important subsystems of the engine. It is the entrance point for air into the engine. In order to increase the power output of the motor, the amount of air flowing into the engine must be maximized. However, the rules for the FSAE competition mandate that the engine must breathe through a 20mm restriction. This constraint poses an immense challenge for the intake system design. The intake system is comprised of many components, which can be seen in Figure 4 (on page 7). These include the following: Throttle body throttle mechanism that regulates the amount of air flow into the engine and thus engine power output Bellmouth helps improve air flow by reducing flow separation and losses 6

15 Restrictor conical duct that leads air from the throttle body to the plenum Plenum collecting chamber for incoming air into the engine Intake runners ducts connecting the plenum to the engine intake ports Figure 4: Components of the Intake System There are three standard intake system designs: a trapezoid style, a log style, and a cone style. The trapezoid style intake, which can be seen in Figure 5 (on page 8), allows air to enter at the top of the trapezoid. Straight intake runners that attach directly to the intake ports on the motor are commonly used. The log style intake, which can be seen in Figure 6 (on page 8), has a side-mounted entrance for air. The runners typically have a simple geometry, but may incorporate bends for packaging or fuel injection purposes. The cone style intake shown in Figure 4, has its air entrance centered about the top of the cone, such that each intake runner receives an equal amount of air. The runners have a complex geometry because they have to start at the base of the plenum and mate with the intake 7

16 ports on the motor. Table 2 lists some of the advantages and disadvantages of each type of system. Trapezoidal intake Figure 5: 2004 Racecar Showing the Trapezoidal Intake System Intake Style Trapezoid Log Cone Figure 6: Log Style Intake System Table 2: Advantages and Disadvantages of the Three Intake System Designs Advantages easy to manufacture good flow distribution to all runners easy to manufacture easy to test different parameters equal flow distribution between all runners Disadvantages difficult to ensure a stiff structure unequal flow distribution between all runners difficult to manufacture complex runner geometries 8

17 3 Literature Review Many technical papers and research books have been written on the topic of intake system design. However, a majority of these resources focus on the design of intake systems for production vehicles or for professional racing applications. The intake system design for an FSAE car is drastically different than for other applications because the engine must breathe through a 20mm restriction. Therefore, the design methodology of conventional intake systems cannot be used in the FSAE series. Since the restrictor design constraint only exists for this Formula series, there are few resources available on the topic of intake system design for FSAE cars. Students on various teams have written technical papers discussing specific intake system designs that they have tested. However, this Formula series does not restrict teams to a specific type of engine, nor does it restrict modifications to the internal components of the engine. These factors, as well as many others, have a great effect on engine performance. Therefore, not all of the information contained within the available resources is useful. However, certain trends that have been determined by other teams may help in the development of new designs. The exhaust system design also affects the output performance of the motor, which means that the intake and exhaust systems have to be designed together to achieve an optimal output. Since the data available in technical papers cannot be used directly in the design of the intake system for the U of T racecars, the aim of this thesis project is to provide relevant information regarding the design of the intake system for future U of T vehicles. 9

18 4 Objectives 4.1 Design Objectives The objective of this thesis project is to develop an optimal intake system design for the U of T FSAE racecar. In order to create an optimal design, the following parameters need to be considered: Increase the overall power output of the engine Produce a broad, flat torque curve between 6,000rpm and 10,000rpm Minimize the overall weight of the system Increase the reliability of the overall system Keep the center of gravity as close to the ground and as close to the middle of the car as possible to reduce the moment of inertia of the overall vehicle Ensure that the system will be easy to service and maintain The 2008 system will be compared to the 2007 system in terms of the aforementioned parameters, in order to determine if an optimal design was achieved. 4.2 Design Constraints There are also several design constraints that must be taken into account when designing the system. These include: A 20mm restriction on the opening of the intake system, mandated by the FSAE rules The overall cost of the 2008 system design must not exceed the cost of the 2007 system The overall system must fit within the constraints of the chassis structure as well as the envelope area specified by the FSAE rules (see Section 1.2.2, page 3) The design, manufacture, and testing required for this project must be completed before March, which is the deadline for this thesis project 10

19 The timeline is the most restrictive design constraint because it limits the amount of development that can be completed before the final system must be manufactured. The packaging requirements will also play a role in the final intake system design, and could affect the maximum plenum volume or runner lengths that can be used for the system. If the cost of the 2008 system is greater than the 2007 system, a cost-benefit analysis will have to be conducted to determine if the system should be manufactured. This is not of great concern because many sponsors of the U of T FSAE Team have generously agreed to donate parts for the 2008 intake system design. 11

20 5 Design Methodology Several design methods were used in an effort to accomplish the goals of this thesis project. These included benchmarking previous systems, creating engine simulations using Ricardo Wave software, conducting physical tests using an engine dyno, choosing the optimal design using a vehicle simulation program, and incorporating several manufacturing methods to create the intake system. 5.1 Benchmarking of Previous Systems Since the U of T FSAE team has been designing and building cars for eleven years, many different systems have been designed and tested and a lot of data has been gathered. This information was used to help develop the design of the 2008 intake system. All three of the common intake styles have been used on previous vehicles. However, a direct comparison of these systems could not be conducted, because the parameters of each system were different and they were all coupled with different exhaust systems. Table 3 lists the characteristics of the intake and exhaust systems from the past four years. The qualitative performance of each of these systems was determined from previous drivers experiences and has been summarized in Table 4 (see page 13). This data, combined with theoretical information on intake system design, allowed for the effects of changes to different system parameters to be hypothesized. The style of plenum to use, the plenum volumes to test, and the lengths of runners to test, were then determined for the 2008 intake system. Table 3: Intake and Exhaust Characteristics of Previous Systems Intake: Runner Length Plenum Volume 3L 1L 1.5L 2L Intake Style Trapezoid Cone Log Log Exhaust: Primary Length

21 Table 4: Summary of Previous Drivers Experiences with Past Intake Systems Intake System Driver Experience 2004 Poor engine output. Resulted in poor on-track performance regardless of the suspension setup 2005 Relied heavily on resonant tuning. Had peak performance at 7,500rpm and 11,500 rpm. Lacked transient response due to fuel mapping issues caused by the peaky performance 2006 Good performance from 6,500rpm to 12,000rpm. Below 6,500rpm, there was a large section where the engine had a negative pressure wave at intake valve close that reduced performance. This made it difficult for corner exiting 2007 Made good low and midrange power, from 5500 rpm to 9500 rpm. Very easy to drive around the track, compared to the previous years. Lacked power above 9,500rpm In recent years the design of the U of T racecars has been evolutionary in nature, instead of revolutionary meaning that the car has been designed based on experience from previous years. Once a system has been optimized to its greatest potential, a new design is created. The 2007 racecar is currently the most optimized car and will therefore be used as a performance benchmark for UT2008. This means that changes to the system parameters will be evaluated based on their performance relative to the 2007 system. This thesis project will be successful if the performance of the 2008 system is superior to that of the 2007 system. 5.2 Ricardo Wave For the first time on the U of T Team, an engine simulation software package called Ricardo Wave was used to model the engine system. This program allows the designer to model and simulate the engine system by specifying dimensional and operational parameters. The purpose of using this software was to increase the efficiency of tuning the intake system, by allowing many design iterations to be tested without manufacturing every design. 13

22 Computer simulations of complex dynamic systems, such as the engine, are advantageous because many design iterations can be performed. However, they are also limited in the sense that they are unable to accurately represent the physical system. A software model approximates certain operating conditions and parameters, which may lead to inaccuracies in the results due to oversimplifications in the model. Ricardo Wave, in particular, is a one dimensional flow analysis, that can provide meaningful trends for engine development, but will not be able to accurately predict the final output of the actual system. Also, since this is the first time that the U of T Team is simulating the engine system, it is important to ensure that the results of the simulation model are correct in terms of predicting the performance trends of the motor. Therefore, some physical testing was necessary to validate the computer simulations. 5.3 Dynamometer An engine dyno was used to conduct the physical testing of the intake system. The purpose of the physical testing was to validate the simulation model in terms of its accuracy of predicting performance trends. In order to maintain low costs and increase the efficiency of the testing process, it was determined that only a few of the designs developed through the simulations would be tested on the dyno. The outputs of the dyno test would then be compared to the outputs of the computer simulation, to determine the validity of the simulations. The 2007 exhaust system was used for both the simulation model and for dyno testing such that the effects of the changing intake system parameters could be determined. The intake and exhaust designs are dependent on each other in terms of overall engine performance so the two systems must be developed together. Therefore, once the trends of the intake system parameters were determined, the exhaust system was modified to determine whether any appreciable gains could be made to the overall system. 14

23 5.4 Manufacturing Methods In order to satisfy the goals of a lightweight design, it was decided to manufacture the 2008 intake system using a combination of aluminum, steel, and nylon, instead of using primarily aluminum, as was done for the 2007 system. The plates of the throttle body were waterjet cut and post machined from 1/ T6 aluminum billet. Most of the small moving components of the throttle body were machined from 4340 steel. The bellmouth was spun from aluminum using a mold. The restrictor-plenum assembly and the intake runners were made from a nylon material (Duraform Polyamide) using a selective laser sintering (SLS) rapid prototyping process. The material properties of the nylon material that was used can be found in Appendix B. The complex geometry of the restrictor-plenum assembly and the intake runners, due to packaging constraints, made the rapid prototyping option the most favourable in terms of manufacturability and time constraints. Also, since the U of T FSAE Team was able to find a rapid prototyping sponsor that was willing to manufacture this system, the cost of this system was no longer an issue. 15

24 6 Initial Intake System Design 6.1 Plenum Style A cone style intake was chosen for UT2008 due to the geometrical symmetry of the design that allows for equal filling of air in all four intake runners. This is an important property that enhances the efficiency of the engine and increases its reliability. Unequal air-fuel mixtures in each engine cylinder can induce unwanted vibration modes in the crankshaft. This is caused by the differences in the work output of each piston which may lead to premature fatigue failure of internal engine components. Although the air flow property of the conical intake design was very appealing, there was a concern that the stiffness of a conical structure would decrease with increasing volume. Therefore, a stiffness comparison of a log style intake (which is considered to be the stiffest structure) and a cone style intake was conducted. This was done through a finite element analysis (FEA) of the deflection of these structures when subject to an external pressure of 14.6psia (atmosphere) and an internal pressure of 0psia (vacuum). The FEA results for the log style intake can be seen in Figures 7 (a) and (b) (see page 17) and the results for the cone style intake can be seen in Figure 8 (see page 17). The volume of the cone was much larger than the volume of the log (which had the same volume as the 2007 intake system) since the volume required for the cone style intake was unknown. The wall thickness and material properties of the two structures were set to be the same. It was found that the maximum deflection in the log style intake in the axial direction was 1.13 x 10-4 in and the maximum deflection in the radial direction was x 10-6 in. The maximum deflection in the cone style intake was 3.73 x 10-4 in. The deflection of the cone style intake was therefore determined to be negligible, so it was decided to use this design for the 2008 intake system. 16

25 Figure 7 (a): FEA of the Log Style Intake for Displacement in the Axial Direction Figure 7 (b): FEA of the Log Style Intake for Displacement in the Radial Direction Figure 8: FEA of the Cone Style Intake for Displacement 17

26 6.2 Modular Intake System Design for Dyno Testing In order to facilitate dyno testing of various intake system parameters, a modular intake system was designed and manufactured from a nylon material using an SLS rapid prototyping process. The requirements for this design included the following: only one intake system could be manufactured due to time and cost constraints the plenum volume had to be easily and quickly modified the intake runner lengths had to be easily and quickly modified Isometric View Exploded View Figure 9: Three Dimensional Model of the Modular Intake System Therefore, a three dimensional modular design was created using Pro Engineer software. Figure 9 shows an isometric view and an exploded view of the modular intake system design. This modular design incorporated several breaks in the system geometry, where extra material could be added. The plenum was divided into two halves, allowing volume changes to be made by adding a straight section of plastic tube. This tube had the same inside diameter as the plenum, to reduce flow losses that occur at abrupt geometrical changes. This same concept was incorporated into the design of the intake runners, to allow the lengths of the intake runners to be changed. The plenum was designed to have the smallest volume required for testing, and the runners were designed to have the 18

27 shortest length required for testing. Aluminum inserts were used to change the intake runner lengths and a plastic insert was used to change the plenum volume. This can be seen in Figure 10, which shows the modular intake system being tested on the dyno. Plenum insert Aluminum insert Figure 10: Modular Intake System being Tested on the Engine Dyno 19

28 7 Testing 7.1 Ricardo Wave Simulation Model Since this was the first time that the U of T FSAE Team was using Ricardo Wave to simulate engine performance, validation of the simulation outputs through physical testing was required to ensure that the simulation model was comparable to the physical model. The first model that was developed for the simulation was of UT2007, because torque curves for this system were available from dyno data collected during the 2007 season. Figure 11 shows a screen shot of the 2007 system that was modeled in Ricardo Wave. Figure 11: Screen Shot of the UT2007 Intake System Model in Ricardo Wave From this Figure it can be seen that the model is quite complex in nature. The software requires the user to input dimensional and operational parameters for each part of the system that is modeled. It also requires a different part to be modeled at each geometrical change. For example, the model of an intake runner with a 60 degree bend radius, as shown in Figure 12 (see page 21), would require three separate parts one part for the straight section before the bend, another part for the bent section, and a third part for the straight section after the bend. The dimensional and operational parameters for each of 20

29 these parts would then have to be specified. This causes the system model to become quite complex. Therefore, many assumptions and simplifications had to be made to the system in order to complete the model. For example, the exhaust primaries were modeled as straight pipes, instead of with multiple bends, as they are on the racecar due to packaging constraints. The assumptions and simplifications that were made to the model did not alter the mechanisms and principles of the engine operation. The geometrical simplifications that were made may have an effect on the overall volumetric efficiency of the engine, but would not affect the output characteristics of the different configurations. Subsequent models were not created due to difficulties experienced with the outputs of the initial model. Section 2 bent section Section 1 straight section before the bend Section 3 straight section after the bend Figure 12: Bent Intake Runner Sections Required for Ricardo Wave Model 7.2 Dynamometer Testing The intake system parameters that were tested on the dyno were determined by examining the performance of past vehicles. Research of intake system designs for high performance systems provided insight into the trends found from changing different intake system parameters. However, it was unknown if these trends would be applicable to a restricted motor. In order to ensure that the dyno was functioning properly and was calibrated correctly, the performance of the dyno had to be measured. This was accomplished by comparing torque curves of the 2007 engine system from data collected last year and data collected 21

30 this year. Therefore, the first test that was performed on the dyno was of the 2007 intake and exhaust system. Figure 13 shows a comparison of the two torque curves. It can be seen from this Figure that the curves are very similar. The profile of the curves is the same, and the torque values are very similar, where the greatest difference between the values is 3%. Therefore, it was determined that the dyno was ready to test the 2008 intake system. The repeatability of the dyno data was important to ensure that the results being collected were accurate. In order to measure the repeatability of the data, two dyno tests were performed for each intake configuration. If the difference between the results was greater than 3%, another test was conducted. If there was still a discrepancy in the data, the dyno was recalibrated and system components, such as spark plugs, were checked to ensure they were still functioning properly. Figure 13: Comparison of Torque Curves for the 2007 Engine System Torque [ft-lbs] torque curve 2008 torque curve RPM 22

31 Since the time constraints to complete this project were very short, it was decided to focus on two parameters for intake testing: the intake runner lengths and the plenum volume. Table 5 lists the various parameters that were tested. Table 5: Intake System Parameters Tested on the Dyno Test Number Plenum Volume Intake Runner Length 1 1.5L L L L L 11.5 All other variables on the motor were kept constant throughout the intake system testing to ensure that the data collected would be comparable. The 2007 exhaust system was used to conduct this testing, to allow the cone style plenum to be compared to the log style plenum. There are many other parts of the engine system that have an effect on the output performance of the motor. Although the focus of this research project was the design of the intake system, one of the main goals was to produce a broad, flat, torque curve throughout the engine operating range. In order to achieve this goal, it was necessary to test other engine parameters to determine if changes to these systems would result in the desired output performance. Therefore, in addition to the intake system parameters that were tested, a different camshaft was tested and several exhaust primary lengths were tested. 7.3 Vehicle Simulation The torque curves produced by the dyno were used to determine the trends produced by changing various intake system parameters. However, in order to determine which intake system was optimal for UT2008, the vehicle performance and the drivability of the systems had to be taken into account. Therefore, a vehicle simulation program was used to help choose the intake system that best achieved the goals for UT

32 The vehicle simulation program was developed by Vince Libertucci, the suspension design leader for the U of T FSAE Team. This program was developed to increase the efficiency of suspension tuning by measuring the effects of different suspension settings in terms of their effects on lap times. Most of the inputs for the simulation are suspension related parameters that affect the handling of the car. However, the simulation requires torque values at various engine speeds, as an input to calculate the lap times. Therefore, the torque curves produced on the dyno from the various intake systems were input into the simulation program. The suspension parameters were held constant and the various torque curves were compared in terms of lap times produced by the simulation. Since the FSAE competition is made up of several different dynamic events that test various properties of the race car, it was important to include all of these aspects in the simulation program in order to properly choose the optimal intake system. Therefore, the track that was used for the simulation incorporated various radius corners and different length straight sections. The lap times at each of these sections of the track were determined, the overall lap time was calculated, and the number of shifts required throughout the lap was noted. This information was used to compare each intake system and to determine which system had the best performance. 24

33 8 Results and Analysis 8.1 Results of the Ricardo Wave Simulation The simulation model of the 2007 engine system was completed and debugged and an output torque curve was generated. Unfortunately, this torque curve, which can be seen in Figure 14, does not provide any meaningful data because it only shows torque values between 0rpm and 5,000rpm. None of the dynamic events in the FSAE competition require the vehicles to operate at such low engine speeds. Therefore, data collected in this operating range is not useful for the design of the intake system. The Ricardo Wave program is capable of creating torque curves throughout a wide range of engine speeds. However, a large investment in time would be required to learn how to complete this function because it is quite complex. It was also determined that the computing time (approximately 24 hours) required to complete a simulation was longer than the time required to complete a test on the dyno (approximately 4 hours). Therefore, due to time constraints, it was decided to continue developing the Ricardo Wave model during the off-season, such that it could be used to develop the 2009 system. Figure 14: Torque Curve Produced in Ricardo Wave for the 2007 Engine System 25

34 8.2 Results of the Dynamometer Testing Important Notes to Consider When Interpreting the Dyno Results The results obtained from the dyno are very sensitive to ambient operating conditions because the torque produced by the motor is affected by the temperature and pressure of the intake air. In general, an engine operates more favorably, i.e. has a greater torque output, in cooler temperatures and at higher pressures. Since all of the dyno testing could not be completed in a single day, nor could the testing be conducted under the same operating conditions, all of the torque values obtained by the dyno had to be corrected to account for the differences in temperature and pressure on each day the dyno was used. Also, since the dyno was run in an indoor environment, the temperature of the air in the room increased throughout each dyno run. Therefore, an air temperature sensor was installed in the plenum and was connected to the ECU, which recorded the inlet air temperature at a rate of 10Hz, or ten samples per second. A weather station was also set up to record the air pressure in the room on a daily basis. The SAE J1349 Standard [3] was used to correct the torque data obtained by the dyno. The formula used for this standard is the following: corrected _ torque = actual _ airtemp( C) actual _ torque * 293K kPa * actual _ pressure Each dyno test was conducted twice to ensure that the data being collected was repeatable within 3%. The torque arm of the dyno was also calibrated each day the dyno was run to prevent any possible errors in the readings. The torque values that were collected were taken at a 100% throttle opening position, meaning that the maximum airflow was being input into the engine. Therefore, the engine was being operated at its full potential such that the maximum torque output at each engine speed could be recorded. This allowed the various intake systems to be compared more accurately, because the performance of each system was at its maximum values. 26

35 In order to determine the trends that were found by changing various intake system parameters, a comparison of several torque curves had to be completed. The following trends were examined in terms of their effect on the torque output of the motor: intake runner length plenum volume camshaft profile exhaust primary length The following naming convention was used to identify the various system configurations that were being tested: X_Y_Z_run # o X = length of the intake runner insert measured in inches 1 insert corresponds to an 8.5 runner length 2 insert corresponds to a 9.5 runner length 4 insert corresponds to an 11.5 runner length o Y = length of the plenum insert measured in inches 0 insert corresponds to a 1.5L plenum volume 4 insert corresponds to a 3L plenum volume o Z = exhaust configuration 06 corresponds to the 2006 exhaust system which has a 20 primary length 07 corresponds to the 2007 exhaust system which has a 26 primary length o # = run number 1 corresponds to the first run 2 corresponds to the second run If the word cali was mentioned in the identification name, it means that a California specification (Cali spec) intake camshaft was used instead of the stock intake camshaft. The exhaust camshaft was never changed. 27

36 8.2.2 Results of Runner Length Testing Figure 15 shows a comparison graph of the torque curves for three different intake runner lengths. The following parameters were constant throughout these tests: 3L plenum volume stock camshafts 26 exhaust primary length Figure 15: Torque Curves for Intake Runner Length Testing Torque [ft-lbs] _4_07_run 1 1_4_07_run 2 2_4_07_run 1 2_4_07_run 2 4_4_07_run 1 4_4_07_run RPM It can be seen from this Figure that as the intake runner length increases, the peak torque value also increases. However, the speed at which this peak torque occurs is lower in the engine operating range compared to where the peak torque values occur on the other systems. In terms of the power output of the motor, which can be seen in Figure 16 (on page 29), the shorter runner length (8.5 ) produces the highest peak power at the highest engine operating speed. 28

37 Figure 16: Power Curves for Intake Runner Length Testing Power [hp] RPM 1_4_07_run 1 2_4_07_run 1 4_4_07_run 2 Table 6 summarizes the trends in the torque and power outputs of the motor for varying intake runner lengths. Table 6: Summary of Torque and Power Output Results for Intake Runner Length Testing Intake Configuration Torque Output Power Output 1_4_07 Highest torque above 8,500rpm Most power above 8,500rpm 2_4_07 Higher torque than 4_4_07 below 6,500rpm and above 8,500rpm Broad power curve from 7,500rpm to 11,500rpm, but lower in amplitude than 1_4_07 4_4_07 Highest peak torque at 7,500rpm, lower torque everywhere else (very peaky) Good from 7,000rpm to 8,500rpm, low everywhere else 29

38 8.2.3 Results of Plenum Volume Testing The torque and power curves that show the effect of changing plenum volume can be seen in Figure 17 and 18 (see page 31), respectively. The following parameters were constant throughout these tests: 9.5 intake runner length stock camshafts 26 exhaust primary length Figure 17: Torque Curves for Plenum Volume Testing Torque [ft-lbs] RPM 2_0_07_run 1 2_0_07_run 2 2_4_07_run 1 2_4_07_run 2 30

39 Figure 18: Power Curves for Plenum Volume Testing Power [hp] RPM 2_0_07_run 1 2_4_07_run 1 From Figure 17 (see page 30), it can be seen that in terms of torque output, the 3L plenum volume has a slightly higher torque value at a lower engine speed compared to the 1.5L plenum volume. However, the smaller plenum volume produces a little more power throughout most of the engine operating speeds, which can be seen in Figure 18. Table 7 summarizes the trends found for changes in plenum volume in terms of the torque and power output of the motor. Table 7: Summary of Torque and Power Output Results for Plenum Volume Testing Intake Configuration Torque Output Power Output 2_0_07 Lower peak value than More power above 7,000rpm 2_4_07, but higher values above 7,000rpm 2_4_07 Higher peak value than 2_0_07 but at a lower rpm Broader curve but lower in amplitude 31

40 8.2.4 Results of Camshaft Testing After conducting the intake runner length and plenum volume tests, it was decided that a different camshaft should be tested. This was partly determined through troubleshooting of the Ricardo Wave simulation, where it was found that small changes made to the valve timing events had a major effect on the torque output of the motor. A few years ago, a Honda CBR600 F4 engine was purchased and it happened to be from California. The strict emissions standards in California resulted in Honda changing the profile of the camshafts for the California motorcycle. Camshaft profiles affect the amount of valve lift and the duration of valve opening. This changes the magnitude of the pressure wave and the frequency at which the pressure wave arrives when the piston is at the intake valve close position. Therefore, camshaft profiles affect the engine speed at which the motor responds to tuning. An approximate comparison of the Cali spec and the stock intake camshaft profiles can be seen in Figure 19. Figure 19: Camshaft Profile Comparison of the Cali Spec and Stock Intake Camshafts Lift (0.001") Cali Stock Crankshaft Degrees This Figure shows that the stock intake camshaft has a much greater valve lift than the Cali spec intake camshaft, and it stays open for a longer duration. The lower lift of the 32

41 Cali spec camshaft provides good cylinder filling of air at low engine speeds, while the higher lift of the stock camshaft provides good cylinder filling of air at high engine speeds. It was unknown how the Cali spec camshaft would respond to the 20mm restrictor design, therefore it was decided to test the Cali spec intake camshaft on the dyno engine. The Cali spec exhaust camshaft could not be tested because it did not have a sensor that was required for the ECU. Figure 20 shows a comparison of the torque curves with the stock intake camshaft and the Cali spec intake camshaft. The following parameters were constant throughout these tests: 8.5 intake runner length 3L plenum volume 20 exhaust primary length Figure 20: Torque Curves for Camshaft Testing torque [ft-lbs] _4_06_stock_run 1 1_4_06_stock_run 2 1_4_06_cali_run 1 1_4_06_cali_run rpm It can be seen from Figure 20 that the Cali spec intake camshaft dramatically improved the torque values in the low engine operating range (between 5,000rpm and 8,500rpm) as well as in the high engine operating range (greater than 9,700rpm). It was hypothesized that the slight dip in the torque curve between 8,700rpm and 9,700rpm may have been caused by the specific intake and exhaust set-up that was being tested. A comparison of 33

42 the power curves, which can be seen in Figure 21, also shows an increase in power in both the low and high ends of the engine operating range. The remarkable increase that can be seen in both the torque and power curves for this particular system configuration, led to the decision that the Cali spec intake camshaft would be used for UT2008. Figure 21: Power Curves for Camshaft Testing Power [hp] RPM 1_4_06_stock_run 1 1_4_06_cali_run 1 Table 8 summarizes the trends found for the different camshafts in terms of the torque and power output of the motor. Table 8: Summary of Torque and Power Output Trends for Camshaft Testing Intake Configuration Torque Output Power Output 1_4_06_stock Higher between 8,500rpm and Overall lower in amplitude 10,500rpm. Low everywhere else 1_4_06_cali Much higher below 8,500rpm and above 10,000rpm Overall higher in amplitude 34

43 8.2.5 Results of Runner Length Testing with the Cali Spec Intake Camshaft Since it was unknown how the Cali spec intake camshaft would affect the trends for intake runner length and plenum volume, these tests were redone on the dyno. The torque curve and power curve for the intake runner trend with the Cali spec intake camshaft can be seen in Figures 22 and 23 (see page 36), respectively. The following parameters were constant throughout these tests: 3L plenum Cali spec intake camshaft 26 exhaust primary length 45 Figure 22: Torque Curves for Intake Runner Length Testing with the Cali Spec Intake Camshaft Torque [ft-lbs] _4_07_cali_run 1 1_4_07_cali_run 2 2_4_07_cali_run 1 2_4_07_cali_run 2 4_4_07_cali_run 1 4_4_07_cali_run RPM 35

44 Figure 23: Power Curves for Intake Runner Length Testing with the Cali Spec Intake Camshaft Power [hp] _4_07_cali_run 1 2_4_07_cali_run 1 4_4_07_cali_run RPM By examining these two graphs, it can be determined that the overall trend for the torque curves is the same for both types of camshafts. The longer runner length (11.5 ) produced the highest torque, but the peak value occurred at a lower engine operating speed than the peak value for the shorter runner lengths. However, the long runner lengths (11.5 ) did not result in higher torque in all areas of the torque curve, unlike for the stock camshaft. Therefore, looking at the power curves, a different trend was found. The medium length runner (9.5 ) had the broadest power curve with the highest horsepower values throughout most of the engine operating range. Therefore, it was decided to use the 9.5 runner lengths for the 2008 intake system. 36

45 Table 9 summarizes the trends in the torque and power outputs of the motor for varying intake runner lengths with the Cali spec intake camshaft. Table 9: Summary of Torque and Power Output Results for Intake Runner Length Testing with the Cali Spec Intake Camshaft Intake Configuration Torque Output Power Output 1_4_07_cali Lowest torque values, but nice Steadily increasing curve profile 2_4_07_cali 4_4_07_cali broad curve profile Higher torque values than 2_4_07 everywhere and also had broad curve profile Highest peak torque, lowest values below 6,000rpm and above 8,000rpm Higher horsepower values than 2_4_07 everywhere and good curve profile Highest values between 7,000rpm and 8,000rpm, lowest values everywhere else Results of Primary Length Testing with the Cali Spec Intake Camshaft One more parameter had to be determined before the 2008 engine system could be finalized. This was the exhaust primary length. The configuration of the exhaust system had been optimized in the 2007 season, however, it was unknown how changes in the primary length would affect the torque and power curves with the Cali spec intake camshaft. Two different exhaust primary lengths were chosen to be compared, based on the availability of resources for the U of T FSAE Team. In 2006, the exhaust system had a 20 primary length and in 2007, the exhaust system had a 26 primary length. These two systems were sufficiently different, such that trends for the torque and power curves could be easily determined. A comparison of the torque and power curves for these two exhaust systems can be seen in Figures 24 (see page 38) and 25 (see page 38), respectively. The following parameters were constant throughout the tests: 11.5 intake runner length 3L plenum volume Cali spec intake camshaft 37

46 Figure 24: Torque Curves for Exhaust Primary Length Testing with the Cali Spec Intake Camshaft Torque [ft-lbs] _4_06_cali_run 1 4_4_06_cali_run 2 4_4_07_ cali_run 1 4_4_07_cali_run RPM 70 Figure 25: Power Curves for Exhaust Primary Length Testing with the Cali Spec Intake Camshaft Power [hp] RPM 4_4_06_cali_run 1 4_4_07_cali_run 1 From Figure 24, it can be seen that the longer exhaust primary length (26 ) produced a higher peak torque value than the shorter exhaust primary length (20 ) that was higher in 38

47 the engine operating range. However, the 2006 exhaust system showed much better torque values in the lower engine operating range. A comparison of the power curves, in Figure 25 (see page 38), showed that the longer exhaust primary length (26 ) produced more power throughout almost the entire engine operating range. Therefore, it was decided that an exhaust system with a 26 primary length should be made for the 2008 racecar. Table 10 summarizes the trends found for the different exhaust primary lengths in terms of the torque and power output of the motor with the Cali spec intake camshaft. Table 10: Summary of Torque and Power Output Results for Exhaust Primary Length Testing with the Cali Spec Intake Camshaft Intake Torque Output Power Output Configuration 4_4_06_cali Higher below 6,500rpm, lower everywhere else Higher below 6,500rpm, lower everywhere else 4_4_07_cali Higher above 6,500 rpm Higher above 6,500 rpm and broader curve profile Comparison of the Modular Intake System and the 2007 Engine System In order to determine if the parameters chosen for the 2008 system resulted in a superior design to the 2007 system, a comparison of the torque and power curves had to be completed. The modular intake system with the following parameters was used for this comparison: 9.5 intake runner length 3L plenum volume Cali spec intake camshaft 26 exhaust primary length The 2007 engine system was tested on the dyno with the Cali spec intake camshaft. Therefore, a comparison was made between the 2007 engine system with the stock intake camshaft, the 2007 engine system with the Cali spec intake camshaft, and the modular intake system with the aforementioned parameters. The torque and power curves for this 39

48 comparison can be seen in Figures 26 and 27, respectively. 44 Figure 26: Comparison of the Torque Curves for the Modular Intake System and the 2007 Engine System Torque [ft-lbs] _07_stock_run 1 07_07_cali run 1 07_07_cali run 2 2_4_07_cali_run 1 2_4_07_cali_run RPM 80 Figure 27: Comparison of the Power Curves for the Modular Intake System and the 2007 Engine System Power [hp] _07_stock_run 1 07_07_cali_run 2 2_4_07_cali_run RPM 40

49 Both of these curves show that the 2007 engine system with the Cali spec intake camshaft is superior to the other systems. It has a higher peak torque value than the other two systems and it makes the most power in the middle and high engine speed ranges. The modular intake system only had better performance in the high engine speed range (above 10,000rpm) compared to the 2007 engine system with the stock intake camshaft. Table 11 summarizes the results from the comparison of the 2007 engine system with the Cali spec intake cam, the 2007 engine system with the stock intake cam, and the modular intake system with the best performing properties. Table 11: Summary of Torque and Power Output Results for the Comparison of the Modular Intake System and the 2007 Engine Systems Intake Torque Output Power Output Configuration 07_07_stock 07_07_cali 2_4_07_cali Higher peak value than 2_4_07_cali, lower values than 2_4_07 above 10,000rpm Highest peak values, high values everywhere Lowest values between 7,500rpm and 10,000rpm, equivalent to 07_07_cali at all other engine speeds Higher than 2_4_07 from 8,000rpm to 10,000rpm, but lower than 2_4_07 above 10,000rpm Highest values above 8,000rpm, good values below 8,000rpm Same as 07_07_cali below 8,000rpm and above 10,000rpm, lowest values in between At this point, it was unknown why the modular intake system did not perform as well as either of the 2007 systems. In theory the cone style plenum should have performed better than the log style plenum due to equal air filling in all of the cylinders. Several possible reasons that may have resulted in the poor performance of the modular intake system were determined: The aggressive divergence angle of the restrictor on the modular intake system may have resulted in high flow losses, limiting the mass air flow into the plenum, thus resulting in poor cylinder filling The bellmouths inside the plenum may have been too close to the walls of the plenum, thus reducing their effectiveness by preventing the air from below the bellmouths to be drawn into the runners 41

50 The throttle body was not aligned exactly with the opening in the restrictor (due to the poor tolerances of the rapid prototype parts), which may have caused less than the maximum amount of air to flow into the intake system at the 100% throttle position Design of the 2008 Intake System All of the possible causes for the lack of performance of the modular intake system were taken into account and a new intake model was designed. If the new intake model worked, it would be used on the 2008 racecar. Therefore, this new model had to be designed such that it could package within the constraints of the 2008 chassis design. Slightly different design parameters were used due to packaging constraints. The final intake system that was designed, which can be seen in Figure 28 (see page 43), had the following characteristics: 10 intake runner lengths 2.75L plenum volume Curved restrictor profile with the same divergence angle as the 2007 restrictor (9 degrees). The curved shape was required to allow the intake system to fit within the envelope area specified by the FSAE rules (see Figure 29 on page 43) Bellmouths in the plenum were raised by 1 compared to the modular intake system design to provide more space between the outer diameter of the bellmouths and the inside of the plenum wall 42

51 Figure 28: Final Intake System Design Figure 29: Packaging of the Intake System within the Envelope of the 2008 Chassis All of the necessary improvements were made to this new design in an effort to optimize the system performance. This new intake design, which will be referred to as the 2008 intake system, was then tested on the dyno, which can be seen in Figure 30 (see page 44). The following tests were performed: 2008 intake system, 20 exhaust primary length, Cali spec intake camshaft 2008 intake system, 26 exhaust primary length, Cali spec intake camshaft 43

52 Figure 30: 2008 Intake System Being Tested on the Engine Dyno The 2008 intake system was compared to the 2007 engine system with the stock intake camshaft and the Cali spec intake camshaft. The torque and power curves for these tests can be seen in Figures 31 (see page 45) and Figure 32 (see page 45), respectively. 44

53 Figure 31: Comparison of Torque Curves for the 2008 Intake System and the 2007 Engine System Torque [ft-lbs] _07_stock 07_07_cali run 1 07_07_cali run 2 08_06_cali_run 1 08_06_cali_run 2 08_07_cali_run 1 08_07_cali_run RPM Figure 32: Comparison of Power Curves for the 2008 Intake System and the 2007 Engine System Power [hp] _07_stock 07_07_cali run 2 08_06_cali_run 1 08_07_cali_run RPM Even with all of the improvements that were made to the design of the 2008 intake system, the above torque and power curves show that the 2007 intake system (with either 45

54 intake camshaft) performs much better than the 2008 intake system throughout the middle engine operating speeds (between 8,000rpm and 10,000rpm). The longer exhaust primary length (26 ) provides better performance for the 2008 intake system than the shorter exhaust primary length (20 ), but neither of these combinations can produce the torque or power that the 2007 systems can achieve. This comparison is a clear indication that the 2007 intake system with the Cali spec intake camshaft should be used for the 2008 racecar. Table 12 summarizes the results of the above comparison tests, in terms of the torque and power outputs of each system. Table 12: Summary of Torque and Power Output Results for the Comparison of the 2008 Intake Systems and the 2007 Engine Systems Intake Torque Output Power Output Configuration 07_07_stock 07_07_cali 08_06_cali 08_07_cali Higher values between 8,000rpm and 10,000rpm compared to the 2008 intake systems Higher values above 8,000rpm than the 2008 intake systems Lowest values from 6,500rpm to 10,000rpm, slightly higher values than 08_07_cali above 10,000rpm Better midrange performance compared to 08_06_cali (between 7,000rpm and 8,500rpm), but overall lower values compared to the 2007 engine systems Higher values between 8,000rpm and 10,000rpm compared to the 2008 intake systems Higher values above 8,000rpm than the 2008 intake systems Lowest values from 7,000rpm to 11,000rpm Same profile as 08_06_cali, but slightly higher values between 7,000rpm to 8,500rpm, but overall lower values compared to the 2007 engine systems It is still unknown why the 2008 intake system design did not perform as well as was originally expected. Several more tests will be conducted to try to determine the possible causes for the poor performance. These will include the following: Increase the diameter of the curved portion of the restrictor on the 2008 intake system. Air flow may be choked in this portion of the restrictor due to the curved profile 46

55 Test different exhaust primary lengths. Only two primary lengths were tested with the 2008 intake system. The 2007 exhaust system had been developed in the 2007 season specifically for the 2007 intake system. However, the focus of the testing for the 2008 season was on the intake system design, so the exhaust testing was not as rigorous Unfortunately, due to time constraints, the results of these tests will not be completed before the deadline for this thesis report. It is hoped that the causes of the lack of performance of the 2008 intake system will be determined before the FSAE competition in May. Since the testing season for the 2008 racecar is about to begin, it has been decided that the 2007 system with the Cali spec intake camshaft will be used for the 2008 season. Further improvements to the 2008 intake system may be used for the 2009 racecar. 8.3 Vehicle Simulation Results Before the 2008 intake system was designed, the vehicle simulation program was used to determine the torque curve that would produce the best lap times under various situations. The track that was used for this simulation consisted of seven corners that were separated by six straight sections. The sizes of the corners and the lengths of the straight sections were based on the track used for the Formula Student competition in Germany. This particular track is ideal for comparing different torque curves because the engine is run through almost its entire operating range. The details of the track that was used are as follows: Corner radii [ft]: 20, 20, 20, 20, 40, 27, 27 Length of corners [ft]: 42, 22, 63, 22, 126, 82, 82 Length of straights [ft]: 40, 97.8, 206, 20, 82, 31 47

56 The minimum corner radius specified in the FSAE rules [1] for a hairpin is 15ft, while the range for sweeping corners is 40ft to 88.6ft. The torque curves for the following system configurations were used for this lap simulation: 8.5 intake runner length, 3L plenum volume, stock camshafts, 20 exhaust primary length (1_4_06) 11.5 intake runner length, 3L plenum volume, stock camshafts, 20 exhaust primary length (4_4_06) 8.5 intake runner length, 3L plenum volume, stock camshafts, 26 exhaust primary length (1_4_07) 11.5 intake runner length, 3L plenum volume, stock camshafts, 26 exhaust primary length (4_4_07) All of the torque curves resulted in equal times in the corners. This is because the speed in the corners is not affected by the engine. The times achieved in each corner can be found in Table 13. Table 13: Corner Lap Times for Each Torque Curve Corner Radii [ft] Corner length [ft] Lap time [sec]

57 Table 14 shows the lap times for each torque curve in the straight sections as well as the total lap times achieved by each torque curve. Table 14: Straight and Total Lap Times for Each Torque Curve Straight Length Lap Times for Each System Configuration [sec] [ft] 1_4_06 4_4_06 1_4_07 4_4_ Total Time From Table 14, it can be seen that the lap times for some of the straight length sections were the same for the different torque curves. This was due to the car being grip limited throughout these sections. It can also be determined that the torque curve for the system configuration 1_4_07, resulted in the fastest total lap time by seconds. Although the laps were very short compared to the tracks that would be used in an FSAE competition, the differences in the lap times of the various system configurations were very small. The repeatability of lap times for an experienced driver is approximately 10%. Therefore, even for a 60 second lap, the vehicle simulation program would result in a maximum of a 1% difference in the lap times for the various torque curves. Therefore, in order to determine which torque curve would produce the optimal lap times, the drivability of the different torque curves had to be analyzed. This was accomplished by determining the number of shifts required in a lap for each torque curve. The results were inconclusive because the lap simulation resulted in the same number of shifts for each lap for all of the torque curves. This occurred because when the racecar exits a corner, the program chooses what gear to use by looking for the gear that will provide the maximum torque at the wheels at its current speed. This always ended up being first gear. In reality, this is very difficult for a driver to achieve. A driver would use second gear instead of first gear in these situations. This is because keeping the racecar in second gear allows the driver to exit the corners easier, instead of trying to keep the car at its acceleration limit throughout the corner. This is a common problem 49

58 that occurs when using a lap simulation program, because the program keeps the car at its limit at all times. The vehicle simulation program was therefore unable to provide a clear solution for the optimal torque curve required to achieve the fastest lap times. Therefore, the optimal intake configuration was determined by examining the torque and power curves produced on the dyno. The intake configuration that best achieved the design objectives for the 2008 intake system design was chosen for use on the 2008 racecar. 50

59 9 Validation of Optimal Design In order to validate that the 2007 engine system with the Cali spec intake camshaft (this system will be referred to as the 2008 optimized intake system) is the optimal design for use on the 2008 racecar, it is important to determine how well this intake design met the design objectives and design constraints outlined in Section 4. The design objectives from Section 4.1 (see page 10) are repeated here for convenience: Increase the overall power output of the engine Produce a broad, flat torque curve between 6,000rpm and 10,000rpm Minimize the overall weight of the system Increase the reliability of the overall system Keep the center of gravity as close to the ground and as close to the middle of the car as possible to reduce the moment of inertia of the overall vehicle Ensure that the system will be easy to service and maintain The 2008 optimized intake system has a greater overall torque output than the 2007 engine system, as can be seen in Figure 31 (see page 45). It also has much higher power values at every engine speed compared to the 2007 engine system, which can be seen in Figure 32 (see page 45). In terms of the overall weight of the system, the 2008 optimized intake system weighs less than the 2007 engine system because the Cali spec intake camshaft is lighter than the stock intake camshaft. The rest of the components (intake and exhaust) weigh the same. Table 15 shows the weight difference between the Cali spec intake camshaft and the stock intake camshaft. Table 15: Weight Comparison of the Cali Spec and Stock Intake Camshafts Cali Spec Intake Camshaft Stock Intake Camshaft Weight [g] The center of gravity as well as the serviceability and maintainability of the 2008 optimized intake system, are equivalent to the 2007 engine system because the components of these two systems are the same, except for the intake camshaft. 51

60 The 2008 optimized intake system was also validated in terms of its ability to meet the design constraints. These were determined in Section 4.2 (see page 10), and are repeated here for convenience: A 20mm restriction on the opening of the intake system, mandated by the FSAE rules The overall cost of the 2008 system design must not exceed the cost of the 2007 system The overall system must package within the constraints of the chassis structure as well as the envelope area specified by the FSAE rules (see Section 1.2.2, page 3) The design, manufacture, and testing required for this project must be completed before March, which is the deadline for the thesis The size of the restriction on the 2008 optimized intake system is the same as on the 2007 engine system. The overall cost of the 2008 optimized intake system is also the same as the 2007 engine system. The modular intake system and the 2008 intake system that were manufactured and tested on the dyno, were donated by Axis Rapid Prototyping, a new sponsor for the U of T FSAE Team. The Cali spec intake camshaft and the exhaust systems used for dyno testing were resources that were already available to the U of T Team. Therefore, no additional costs were incurred through the design, development, and manufacture of the 2008 optimized intake system. The one area of concern was how well the 2008 optimized intake system would package within the envelope area specified by the FSAE rules. This was because the envelope area on the 2008 chassis was much smaller on the 2007 chassis because the height of the main roll hoop was lowered by three inches. Therefore, the packaging of the 2008 optimized intake system was checked using a three-dimensional solid model, which can be seen in Figure 33 (see page 53). It can be seen from this Figure that this intake system fits within the constraints of the envelope. 52

61 Figure 33: Packaging of the 2008 Optimized Intake System within the Envelope of the 2008 Chassis Since the 2008 optimized intake system meets all of the design objectives while taking into account all of the design constraints, it can be determined that this is the most optimized intake system design. Therefore, the goals of the thesis project have been accomplished. 53

62 10 Conclusion The purpose of this thesis project was to develop an optimized intake system design to be used on the 2008 U of T FSAE racecar. The goal was to better understand the effect of changing various intake system parameters on the torque and power outputs of the motor. The results of the computer simulations, using Ricardo Wave and the vehicle simulation program, could not be used to determine an optimal intake system design because they were unable to make a significant distinction between different intake system configurations. Therefore, the optimal design was determined through the examination and analysis of the torque and power outputs of various intake system designs, produced on the engine dyno. Two different cone style intake systems were designed, manufactured, and tested on the dyno. These systems were tested with various combinations of intake, camshaft, and exhaust parameters. The result was that the performance of the cone style intake systems was inferior to the log style intake system used on UT2007. This was not the expected result. However, the reasons for the poor performance of the cone style intake systems were not investigated due to the time constraints for this project. The performance of the 2007 engine system was optimized through the use of the Cali spec intake camshaft. This system produced better torque and power outputs throughout the entire engine operating range compared to the 2007 engine system with the stock intake camshaft, used on UT2007. Therefore, although the cone style intake system did not perform as well as expected, the design objectives of the thesis project were achieved. Therefore, the thesis project was a success in terms of optimizing the intake system design for the 2008 U of T FSAE racecar. 54

63 11 References [1] SAE Collegiate Design Series: Formula SAE Formula SAE Rules. SAE International. pp Accessed: February 20, [2] Land and Sea. January 25, DYNOmite Dynomometer. Accessed: March 5, [3] SAE International SAE J1349 Certified Power. Accessed: January 7, [4] 3D Systems Corporation. December 10, Duraform PA Plastic: Technical Data. Accessed: February 10, DuraForm_PA_plastic_1207.pdf 55

64 Appendix A Formula SAE Rules for Powertrain Design [1] 3.5 Powertrain Engine and Drivetrain Engine Limitations The engine(s) used to power the car must be four-stroke piston engine(s) with a displacement not exceeding 610 cc per cycle. The engine can be modified within the restrictions of the rules. If more than one engine is used, the total displacement cannot exceed 610 cc and the air for all engines must pass through a single air intake restrictor (see , Intake System Restrictor. ) Hybrid powertrains utilizing on-board energy storage are not allowed Engine Inspection The organizer will measure or tear down a substantial number of engines to confirm conformance to the rules. The initial measurement will be made externally with a measurement accuracy of one (1) percent. When installed to and coaxially with spark plug hole, the measurement tool has dimensions of 381 mm (15 inches) long and 30 mm (1.2 inches) diameter. Teams may choose to design in access space for this tool above each spark plug hole to reduce time should their vehicle be inspected Transmission and Drive Any transmission and drivetrain may be used Drive Train Shields and Guards Exposed high-speed equipment, such as torque converters, clutches, belt drives and clutch drives, must be fitted with scatter shields in case of failure. Scatter shields for chains or belts must not be made of perforated material. A. Chain drive - Scatter shields for chains must be made of at least 2.66 mm (0.105 inch) steel (no alternatives are allowed), and have a minimum width equal to three (3) times the width of the chain. B. Belt drive - Scatter shields for belts must be made from at least 3.0 mm (0.120 inch) Aluminum Alloy 6061-T6, and have a minimum width that is equal to the belt width plus 35% on each side of the belt (1.7 times the width of the belt). C. Attachment Fasteners - All fasteners attaching scatter shields and guards must be a minimum 6mm grade M8.8 (1/4 inch SAE grade 5). D. Attached shields and guards must be mounted so that they remain laterally aligned with the chain or belt under all conditions. E. Finger Guards Finger guards may be made of lighter material System Sealing The engine and transmission must be sealed to prevent leakage. Separate catch cans must be employed to retain fluids from any vents for the coolant system or the crankcase or engine lubrication system. Each catch-can must have a minimum volume of ten (10) percent of the fluid being contained or 0.9 liter (one U.S. quart) whichever is greater. 56

65 Catch cans must be capable of containing boiling water without deformation, and be located rearwards of the firewall below driver s shoulder level. They must have a vent with a minimum diameter of 3 mm (1/8 inch) with the vent pointing away from the driver. Any crankcase or engine lubrication vent lines routed to the intake system must be connected upstream of the intake system restrictor Coolant Fluid Limitations Water-cooled engines must only use plain water, or water with cooling system rust and corrosion inhibitor at no more than liters per liter of plain water. Glycol based antifreeze or water pump lubricants of any kind are strictly prohibited Starter Each car must be equipped with an on-board starter, and be able to start without any outside assistance at any time during the competition Fuels The basic fuel available at competitions in the Formula SAE Series is unleaded gasoline with an octane rating of 93 (R+M)/2 (approximately 98 RON). Other fuels may be available at the discretion of the organizing body. Unless otherwise announced by the individual organizing body, the fuel at competitions in the Formula SAE Series will be provided by the organizer. During all performance events the cars must be operated with the fuels provided by the organizer at the competition. Nothing may be added to the provided fuels. This prohibition includes nitrous oxide or any other oxidizing agent. Teams are advised that the fuel supplied in the United States is subject to various federal and state regulations and may contain up to ten percent (10%) ethanol. The exact chemical composition and physical characteristics of the available fuel may not be known prior to the competition. Consult the individual competition websites for fuel types and other information Fuel Temperature Changes Prohibited The temperature of fuel introduced into the fuel system may not be changed with the intent to improve calculated fuel economy Fuel Additives Prohibited No agents other than fuel (gasoline or E85), and air may be induced into the combustion chamber. Non-adherence to this rule will be reason for disqualification. Officials have the right to inspect the oil Fuel System Fuel Tank Size Limit Any size fuel tank may be used. The fuel system must have a provision for emptying the fuel tank if required Filler Neck & Sight Tube All fuel tanks must have a filler neck: (a) at least 38 mm (1.5 inches) diameter, (b) at least 125 mm (4.9 inches) vertical height and (c) angled at no more than 45 degrees (45 ) from the vertical. The 125 mm of vertical height must be above the top level of the tank, and 57

66 must be accompanied by a clear fuel resistant sight tube for reading fuel level (figure 7). The sight tube must have at least 75 mm (3 inches) of vertical height and a minimum inside diameter of 6 mm (0.25 inches). The sight tube must not run below the top surface of the fuel tank. A clear filler tube may be used, subject to approval by the Rules Committee or technical inspectors at the event Fuel Level Line A permanent, non-moveable fuel level line must be located between 12.7 mm and 25.4 mm (0.5 inch and 1 inch) below the top of the sight tube. This line will be used as the fill line for Tilt Test ( and 4.2.3), and before and after the Endurance Test to measure the amount of fuel used during the Endurance Event. The sight tube and fuel level line must be clearly visible to an individual filling the tank Tank Filling Requirement The tank must be capable of being filled to capacity without manipulating the tank or vehicle in any way (shaking vehicle, etc.) Spillage Prevention The fuel system must be designed such that the spillage during refueling cannot contact the driver position, exhaust system, hot engine parts, or the ignition system. Belly pans must be vented to prevent accumulation of fuel Venting Systems The fuel tank and carburetor venting systems must be designed such that fuel cannot spill during hard cornering or acceleration. This is a concern since motorcycle carburetors normally are not designed for lateral accelerations. All fuel vent lines must be equipped with a check valve to prevent fuel leakage when the tank is inverted. All fuel vent lines must exit outside the bodywork. 58

67 Tilt Test-Fuel and Fluids During technical inspection, the car must be capable of being tilted to a 45 degree (45 ) angle without leaking fuel or fluid of any type. The tilt test will be conducted with the vehicle containing the maximum amount of fluids it will carry during any test or event Fuel Lines, Line Attachment and Protection Plastic fuel lines between the fuel tank and the engine (supply and return) are prohibited. If rubber fuel line or hose is used, the components over which the hose is clamped must have annular bulb or barbed fittings to retain the hose. Also, clamps specifically designed for fuel lines must be used. These clamps have three (3) important features, (i) a full 360 degree (360 ) wrap, (ii) a nut and bolt system for tightening, and (iii) rolled edges to prevent the clamp cutting into the hose. Worm gear type hose clamps are not approved for use on any fuel line. Fuel lines must be securely attached to the vehicle and/or engine. All fuel lines must be shielded from possible rotating equipment failure or collision damage Fuel Injection System Requirement The following requirements apply to fuel injection systems. A. Fuel Lines Flexible fuel lines must be either (i) metal braided hose with either crimped-on or reusable, threaded fittings, or (ii) reinforced rubber hose with some form of abrasion resistant protection with fuel line clamps per Note: Hose clamps over metal braided hose will not be accepted. B. Fuel Rail The fuel rail must be securely attached to the engine cylinder block, cylinder head, or intake manifold with brackets and mechanical fasteners. This precludes the use of hose clamps, plastic ties, or safety wire. C. Intake Manifold The intake manifold must be securely attached to the engine block or cylinder head with brackets and mechanical fasteners. This precludes the use of hose clamps, plastic ties, or safety wires. The use of rubber bushings or hose is acceptable for creating and sealing air passages, but is not considered a structural attachment Air Intake and Fuel System Location Requirements All parts of the fuel storage and supply system, and all parts of the engine air and fuel control systems (including the throttle or carburetor, and the complete air intake system, including the air cleaner and any air boxes) must lie within the surface defined by the top of the roll bar and the outside edge of the four tires (see figure 8). All fuel tanks must be shielded from side impact collisions. Any fuel tank which is located outside the Side Impact Structure required by 3.3.8, must be shielded by structure built to A firewall must also be incorporated, per section Any portion of the air intake system that is less than 350 mm (13.8 inches) above the ground must be shielded by structure built to

68 3.5.4 Throttle, Throttle Actuation and Intake Restrictor Carburetor/Throttle Body Required The car must be equipped with a carburetor or throttle body. The carburetor or throttle body may be of any size or design Throttle Actuation The throttle must be actuated mechanically, i.e. via a cable or a rod system. The use of electronic throttle control (ETC) or drive-by-wire is prohibited. The throttle cable or rod must have smooth operation, and must not have the possibility of binding or sticking. The throttle actuation system must use at least two (2) return springs located at the throttle body, so that the failure of any component of the throttle system will not prevent the throttle returning to the closed position. Note: Throttle Position Sensors (TPS) are NOT acceptable as return springs. Throttle cables must be at least 50.8 mm (2 inches) from any exhaust system component and out of the exhaust stream. A positive pedal stop must be incorporated on the throttle pedal to prevent over stressing the throttle cable or actuation system. The use of a pushpull type throttle cable with a throttle pedal that is capable of forcing the throttle closed (e.g. toe strap) is recommended Intake System Restrictor In order to limit the power capability from the engine, a single circular restrictor must be placed in the intake system between the throttle and the engine and all engine airflow must pass through the restrictor. Any device that has the ability to throttle the engine downstream of the restrictor is prohibited. The maximum restrictor diameters are: - Gasoline fueled cars mm ( inch) - E-85 fueled cars 19.0 mm ( inch) The restrictor must be located to facilitate measurement during the inspection process. The circular restricting cross section may NOT be movable or flexible in any way, e.g. the restrictor may not be part of the movable portion of a barrel throttle body. If more than one engine is used, the intake air for all engines must pass through the one restrictor. 60

69 Turbochargers & Superchargers Turbochargers or superchargers are allowed if the competition team designs the application. Engines that have been designed for and originally come equipped with a turbocharger are not allowed to compete with the turbo installed. The restrictor must be placed upstream of the compressor but after the carburetor or throttle valve. Thus, the only sequence allowed is throttle, restrictor, compressor, engine. The intake air may be cooled with an intercooler (a charge air cooler). Only ambient air may be used to remove heat from the intercooler system. Air-to-air and water-to air intercoolers are permitted. The coolant of a water-to-air intercooler system must comply with Rule Muffler and Exhaust System Muffler The car must be equipped with a muffler in the exhaust system to reduce the noise to an acceptable level Exhaust Outlet The exhaust must be routed so that the driver is not subjected to fumes at any speed considering the draft of the car. The exhaust outlet(s) must not extend more than 60 cm (23.6 inches) behind the centerline of the rear axle, and shall be no more than 60 cm (23.6 inches) above the ground. Any exhaust components (headers, mufflers, etc.) that protrude from the side of the body in front of the main roll hoop must be shielded to prevent contact by persons approaching the car or a driver exiting the car Noise A. Sound Measuring Procedure The sound level will be measured during a static test. Measurements will be made with a free-field microphone placed free from obstructions at the exhaust outlet level, 0.5 m (19.68 inches) from the end of the exhaust outlet, at an angle of 45 degrees (45 ) with the outlet in the horizontal plane. The test will be run with the gearbox in neutral at the engine speed defined below. Where more than one exhaust outlet is present, the test will be repeated for each exhaust and the highest reading will be used. The car must be compliant at all engine speeds up to the test speed defined below. B. Test Speeds The test speed for a given engine will be the engine speed that corresponds to an average piston speed of m/min (3,000 ft/min) for automotive or motorcycle engines, and m/min (2,400 ft/min) for industrial engines. The calculated speed will be rounded to the nearest 500 rpm. The test speeds for typical engines will be published by the organizers. The definition of an industrial engine is that used in Rule To have an engine classified as an industrial engine, approval must be obtained from organizers prior to the Competition. C. Maximum Sound Level The maximum permitted sound level is 110 dba, fast weighting. D. Sound Level Re-testing If a car fails the noise test, it will be withheld from the competition until it has been modified and re-passes the noise test. 61

70 Appendix B Material Properties of Duraform Polyamide [4] 62