Phase 4 Report MEEG401: Team FSAE Suspension

Similar documents
SAE Mini BAJA: Suspension and Steering

Design and Analysis of suspension system components

A double-wishbone type suspension is used in the front. A multi-link type suspension is used in the rear. Tread* mm (in.) 1560 (61.

SAE Mini BAJA: Suspension and Steering

PROJECT IDEA SUBMISSION

Suspension Analyzer Full Vehicle Version

Design and Integration of Suspension, Brake and Steering Systems for a Formula SAE Race Car

SUMMARY OF STANDARD K&C TESTS AND REPORTED RESULTS

Wheel Alignment Defined

BIG BAR SOFT SPRING SET UP SECRETS

MM Adjustable IRS Tie-Rod (MMIRSTR-2)

The g-force Family Control Manual

Multilink IRS Setup & Assembly Manual

The g-force Family Control Manual

Kinematic Analysis of Roll Motion for a Strut/SLA Suspension System Yung Chang Chen, Po Yi Tsai, I An Lai

First, check and record the camber and caster readings, they will be adjusted later.

ME 455 Lecture Ideas, Fall 2010

Tech Tip: Trackside Tire Data

QUICKTRICK PRODUCT IN- STRUCTIONS ALIGNMENT SIMPLE SOLUTIONS. Mr & Mrs QuickTrick. Made in the USA

2. Remove front wheels.

INSTALLATION GUIDE Avenger 4-Link, Scalloped for 1-5/8 Clips and Chassis

05-07 F /2 SUSPENSION KIT

What s in the Box? AMT Motorsport C5/C6 Corvette Steel Frame Camber Kit User s Guide

Cane Creek Double Barrel Instructions

INSTALLATION INSTRUCTIONS 64 ½ - 70 SUPERRIDE II INDEPENDENT FRONT SUSPENSION BX-350 FOR COYOTE AND MOD ENGINES

University of Wisconsin-Platteville Formula SAE Design Report

Team FSAE Powertrain

CHASSIS DYNAMICS TABLE OF CONTENTS A. DRIVER / CREW CHIEF COMMUNICATION I. CREW CHIEF COMMUNICATION RESPONSIBILITIES

INSTRUCTIONS FOR STRUT FRONT, 4-LINK REAR, ROADSTER CHASSIS

PRESEASON CHASSIS SETUP TIPS

GENERAL INFORMATION. Wheel Alignment Theory & Operation

Setup Guide and Chassis Tuning Tips (simple version) By Jim Daniels

Next, chase the threads in the lower A-arm mounts with the 5/8-18 tap and blowout any remaining particles.

MM Caster/Camber Plates, (MMCC7989)

Torque steer effects resulting from tyre aligning torque Effect of kinematics and elastokinematics

ROLL CENTER You can adjust the front and rear roll centers of the XB8 by changing the mounting locations of various components.

Newsletter November This month CFS10. Engine. Body. Welcome CFS10 p.1 CFS10 West p.4 What now? p.5 Interested? p.5

AMT Motorsport C7 Corvette Camber Kit User s Guide. 8 Upper Control Arm Studs and hardware for rear upper control arm adjustments

Chrysler A-Body Tubular A-Arms Installation Instructions A-ARM INSTALLATION

CONTENTS. All texts and images contained within this set-up book are copyright by XRAY. All rights reserved. XRAY

COWBOY MOTORSPORTS SENIOR DESIGN Scott Dick Garrett Dollins Logan Gary

What s in the Box? AMT Motorsport C5/C6 Corvette Steel Frame Camber Kit User s Guide

AEV30308AA Last Updated: 05/31/18. 4 DUALSPORT sc SUSPENSION system for RAM 1500 air ride standard and rebel INSTALLATION GUIDE

7211 A-ARM FRONT, 4-LINK, 3 X 2 FRAME, INTERMEDIATE, ELIMINATOR CHASSIS ITEM QTY SIZE/PART NO. TUBE CODE DESCRIPTION

STUDY OF ROLL CENTER SAURABH SINGH *, SAGAR SAHU ** ABSTRACT

1. SPECIFICATIONS 2. WHEEL ALIGNMENT Front Suspension. (gas type) Rear Suspension. (gas type)

Next, set the bar level and tighten it down. Do this on both the driver and passenger sides.

Detroit Speed, Inc. QUADRA Link Rear Suspension Chevy II P/N:

RHINO SUSPENSION SYSTEM INSTALLATION INSTRUCTIONS

Basic Wheel Alignment Techniques

COMPETITION ENGINEERING. 80 Carter Drive P.O. Box 1470 Guilford, CT Phone: (203) Fax: (203)

1. SPECIFICATIONS 2. WHEEL ALIGNMENT

Thanks for Ordering The Kawasaki KLX Adjustable Lowering Kit From

2012 Dalhousie University Formula SAE Design Report

For Technical Assistance, call Competition Engineering's Tech Line at (203) , 8:30am-5:00pm Eastern Time COMPETITION ENGINEERING

Analysis and control of vehicle steering wheel angular vibrations

Design of Formula SAE Suspension

White Paper: The Physics of Braking Systems

INSTRUCTIONS FOR STRUT FRONT, 4-LINK REAR, 1 5/8 FRAME, FULL SIZE, 4130 ELIMINATOR CHASSIS

Introduction: Problem statement

How to Set the Alignment on Ford Mustangs

University of Alberta Design Report

Design, Modelling & Analysis of Double Wishbone Suspension System

2011+ Adjustable Tie-rod Ends (Mm5TR-2)

FSAE SUSPENSION SYSTEM

INSTALLATION INSTRUCTIONS P/N S: C2091 & C2093 SLIDE-A-LINK

X4-X7 Hyper 600cc Chassis Setup Guide

Structural Analysis of Student Formula Race Car Chassis

ASME Human Powered Vehicle

2. MEASURE VEHICLE HEIGHT. (b) Measure the vehicle height. Measurement points: C: Ground clearance of front wheel center

STAGE 1 COIL-OVER SUSPENSION KIT INSTALLATION INSTRUCTIONS PART NUMBER D APPLICATION: 2016 M2 (F87 without EDC)

94-96 Impala SS/ B-Body Rear Coilover Conversion Kit

OTK CHASSIS- SET UP GUIDE

SUSPENSION 2-1 SUSPENSION TABLE OF CONTENTS

7256 INSTRUCTIONS FOR ELIMINATOR II A-ARM FRONT, 4-LINK REAR, MILD STEEL, INTERMEDIATE, SERIES CHASSIS

7333 INSTRUCTIONS FOR MILD STEEL A-ARM AVENGER CHASSIS MUSTANG

Installing and Squaring the Rear Axle

COMPETITION ENGINEERING. 80 Carter Drive P.O. Box 1470 Guilford, CT Phone: (203) Fax: (203)

DESIGN AND ANALYSIS OF PUSH ROD ROCKER ARM SUSPENSION USING MONO SPRING

Designing and Hard Point Optimization of Suspension System of a Three-Wheel Hybrid Vehicle

Hub Stands -- VERSION 5.0

2010 Sponsorship Information Package

ATASA 5 th. Wheel Alignment. Please Read The Summary. ATASA 5 TH Study Guide Chapter 47 Pages: Wheel Alignment 64 Points

'64-72 Chevelle/ A Body Rear Coilover Conversion Kit

Development of a Multibody Systems Model for Investigation of the Effects of Hybrid Electric Vehicle Powertrains on Vehicle Dynamics.

Variable Valve Drive From the Concept to Series Approval

2017 Baja SAE Competition

Jim Kasprzak 36 years racing experience Developed 7-post testing for GM Racing Currently Race Engineer for SRT Viper Expertise includes:

1204AA Ford Mustang Double Adjustable Trailing Arms

2011 F250 6 SUSPENSION KIT

05-07 F250 6 SUSPENSION KIT

HEIDTS SUPERIDE IRS INSTALLATION INSTRUCTIONS INDEPENDENT REAR SUSPENSION

Adjustable Tie-rod Ends (Mm5TR-1)

SPMM OUTLINE SPECIFICATION - SP20016 issue 2 WHAT IS THE SPMM 5000?

STAGE 1 COIL-OVER SUSPENSION KIT INSTALLATION INSTRUCTIONS PART NUMBER R APPLICATION: M3 (E90, E92, E93; non-edc only)

Sequoia power steering rack service Match-mounting wheels and tires Oxygen sensor circuit diagnosis

Installation Guide for TeraFlex 3-inch Lift for Grand Cherokees (3Z)

3 AND 5 LINK INSTALLATION MANUAL

Typical mounting of a dial indicator for a radial check. Moog Automotive, Inc.

Detroit Speed, Inc. QUADRA Link Rear Suspension Nova P/N:

Transcription:

Phase 4 Report MEEG401: Team FSAE Suspension December 10, 2010

2 Introduction: For Team FSAE Suspension, phase 4 was largely spent in the student machine shop. The final design selected during phase 2, and developed in detail during phase 3, was manufactured and installed. Many hours were spent in the manufacture of the components of the proposed suspension system. The result of this semester s work is the research, concept generation, concept selection, and concept implementation of a successful unequal a-arm independent suspension for the 2010 UD FSAE race car. The final touches are still being made to the assemblies, but the components have been manufactured, and for the most part installed on the new car s chassis, and will shortly be subjected to testing and performance evaluation. The completed design package includes Uprights, A-Arms, Push Rods, Tie Rods, Rockers, Shock Absorbers, Sway Bars, and all of the clevises, mounting tabs and hardware needed to assemble and test the suspension. During the final few days remaining before the deliverables are due to the sponsor, the team intends to continue with the testing further defined in this report, and will pursue ongoing work with the FSAE club during the winter session and spring semester to continue to test and tune the car s suspension. Final Design Description: Suspension Style: The unequal a-arm suspension was chosen by the suspension team as the design concept to be pursued during phase two. The A-arm configuration meets the design requirements for structural integrity and suspension travel, and benchmarking specific to FSAE indicated that the Unequal A-arm design was by far the most common design in FSAE application. The chosen design concept utilizes similar components for front, back, top and bottom control arms, which will provide a high level of standardization of parts and components, and the similarity to the prior car's design will enable easy use of components such as shocks and springs from previous cars. Each control arm has been manufactured with a few simple welding and machining operations, and the final dimensions of the upper and lower arms across all four corners of the car have been standardized so that two configurations, top a-arm and bottom a-arm, are similar. Other parts such as rockers, pushrods and clevises have been made as uniform as possible to keep the number of discrete parts as low as possible. The unequal a-arm suspension system is comprised of upper and lower V shaped frames attached to the chassis in four specified suspension mounts with spherical bearings to provide one degree of freedom. Each arm provides an attachment point for the wheel upright that is fixed in two dimensions (laterally and longitudinally to the chassis) and free to move in the third

3 dimension (vertically). The upper and lower arms are designed with different lengths and mount points, with the shorter arm installed above the longer arm. This geometry provides a change in camber as the suspension components move through the range of travel. The design lengths have been optimized through the kinematic analysis to provide the desired cambering behavior. The suspension arms are supported by a push rod assembly that resists the motion associated with the suspension travel, and supports the weight of the car in its neutral position. The push rod attaches at the base of the wheel upright, and travels diagonally upward to an attachment to a rocker. The rocker is affixed to the chassis, and transfers the force in the push rod to a shock absorber and spring assembly that both dampens, and supports the travel of the suspension system, and to a pushrod that actuates the sway bars during body roll. Figure 1: Upper and Lower A-arms Adjustability: One of the most important details that the sponsor emphasized for the design of the suspension was adjustability. After careful consideration of the advantages and disadvantages of each of the adjustability concepts discussed in the previous section of this report, three stood out as the most promising: the threaded chassis mount, slotted chassis clevis, and upright shims. These three ideas work in conjunction in the suspension system to achieve the desired degree of adjustability. The merits of the three concepts are discussed below. Threaded Rod End This concept achieves adjustability in the a-arm length. The rod end is easily accessed by removing a bolt and screwing it in or out to adjust the a-arm length. Additionally, the same process can used to fine tune the caster angle. This concept did the best job of avoiding weakness in the a-arm. It keeps the a-arm solid, with only a threaded insert at the end. This design moves the weakness of the suspension to the bolts. If the bolts should fail, they are easily replaced and we save the more important components like the a-arms. Team FSAE suspension is confident that this will not become a problem. This concept is one of the easiest

4 to adjust, requiring only the loosening of a bolt. The idea can also be standardized to be compatible with all of the mount points, significantly reducing the number of drawings that will have to be developed. Additionally, once the optimal geometry is determined through testing on the car, the mount can easily be replaced with a rigid clevis mount, improving the strength and reliability of the joint. spherical rod end mounts to chassis Locknut Threaded A-arm Figure 2: Threaded Rod End Chassis Pickup Point Upright Shims To achieve the desired camber adjustability, it was decided that the most effective way would be to add shims in between the clevis and the upright. Each shim thickness corresponds to specific degrees of camber change. It is very easy to implement by simply loosening the bolts that attach the clevis to the upright and slipping a shim in between them. This concept is commonly practiced and the most reliable, so it has been implemented in our design. shim upright a-arm clevis Figure 3: Upright Shims

5 Chassis Pickup Bushings This design simply and efficiently achieves anti-squat and anti-dive through vertical adjustment of the pickup points. This concept was chosen for its effectiveness and simplicity. It cut down on the number of the required parts, and fully achieved the necessary purpose. This adjustable mount will only be incorporated on the bottom pickup point nearest to the end of the car lengthwise, on each of the four corners. Multiple sets of different length bushings have been manufactured, and these sets will be switched in and out of the pickup points until the optimal anti-squat and anti-dive characteristics are achieved. The method of adjustment can be replaced by a non-adjustable mount on cars for future years once the optimal geometry is determined through this year s testing. bushing bushing Figure 4: Adjustable Chassis Pickup Point Hubs and Uprights: The front upright design has been reused as much as is possible for this project for both front and rear upright applications. The front upright design from the past car was deemed acceptable by the project sponsor and advisor, and upright design was determined to be outside the scope of this project, however; the scope of the project does dictate some changes with respect to the rear hub and breaking system. The former car used a different hub than the one that the power train team has chosen, and utilized a single axle mounted rear brake system. The new suspension design includes rear uprights that accept the hub from the drive axle, as well as provide caliper mounts for a wheel mounted brake system. This accommodation

6 has been made by modifying the front upright design (that already includes caliper mounts) to receive the rear hub. The specific hub and bearing configuration has been chosen by the drivetrain team, and the suspension team has accommodated by modifying the existing upright design. Shocks and Springs: Figure 5: Front and Rear uprights As per the customer and advisor recommendations, the Fox shocks already owned by the UD FSAE club have been utilized in the design. Springs of varying stiffnesses are readily available for these shocks, so the appropriate springs have been selected and purchased. Using vehicle dynamics theory, as researched from an authoritative text on the topic, Race Car Vehicle Dynamics, Milliken and Milliken, the desired ride and roll rates were found. Ride rate is defined as the force required to make the wheel travel upwards per unit of distance. It defines not only the travel of the suspension and the vertical movement of the chassis, but also the roll characteristics of the car through the associated spring rate. As the car rolls, it causes the outside wheel to travel upwards due to the centrifugal moment about the roll center. The stiffness of the springs directly affects how much the car can roll.

7 Roll rate is the torque required to rotate the car about the roll center per unit angle. Thus, the roll rate is defined by the centrifugal moment about the roll center in cornering situations. Two components define the roll rate: the spring stiffness and the sway bars. The spring stiffness contributes to the roll stiffness as described above. The sway bars contribute to the roll rate by utilizing a torsional spring. The spring links to the rockers on either side of the car, and resists travel of the wheels in opposite directions. The desired ride rate defines the spring stiffness, and the remaining roll stiffness desired is contributed by the sway bars. Using vehicle dynamics theory, the optimal ride and roll rates were calculated. After this, the geometry of the suspension was used to determine the necessary spring stiffness. From this, the stiffness of the sway bars was calculated and translated into the design presented in this report. The values are shown below. Table 1: Ride, Roll, and Spring Rates Front Ride Rate 148.4 lbs/in Rear Ride Rate 146 lbs/in Front Roll Rate 18750 lb ft/rad Rear Roll Rate 20875 lb ft/rad Front Spring Rate 250 lbs/in Rear Spring Rate 250 lbs/in Front Sway Bar Rate 50 lb ft/deg Rear Sway Bar Rate 50 lb ft/deg Performance Evaluation and Test Plan: In order to gauge the success of the design at meeting the target values established, it was necessary for a test plan to be developed. This plan needed to successfully measure all of the metrics, so that the actual value for each metric could be compared to the target value, and thus measure the success of the project solution. The methods of evaluating performance are summarized in the Table 2, and discussed in detail in the following pages. Any actual values that are not yet filled in will be measured by the final assembly of the car, and this document duly updated. Table 2: Metrics, Target Values, and Methods of Evaluation Metric Target Value Actual Value Evaluation Method Wheelbase 61 Final Car Assembly Front Track 50 Final Car Assembly Rear Track 48 Final Car Assembly Anti-Dive and Anti-Squat + 1" vertical movement of bottom pickup points closest -1 vertical movement of bottom pickup points Suspension Assembly and Measurement

8 to CoG of car farthest from CoG of car Roll Center Stable, < 1" vertical movement over 1.5 ---- Suspension Assembly and Measurement deflection in roll, ~3 height at neutral Scrub Radius Minimize, <1 Suspension Assembly and Measurement Camber -2: static, maintained over 0.5 deflection in roll ---- Suspension Assembly and Measurement Kingpin Angle 0-5 degrees 0 degrees Suspension Assembly and Measurement Caster Angle 0-5 degrees Suspension Assembly and Measurement Center of Gravity Longitudinal: centered Lateral: centered Height: 10 F/R Weight Distribution 50/50 Spring Stiffness TBD 250 lbs/in Damped Frequency TBD 3 3.5 Hz # Tools to Adjust 3 3 (2 x 7/16 wrenches and 1 x 9/16 wrench) Suspension Assembly and Measurement Adjustments easy to access Yes Yes (judged by all team members) Suspension Assembly and Measurement Camber and Toe adjustment without Yes Yes Suspension Assembly and Measurement disconnection of parts Material Strength Factor of safety >2 for range of normal operation FOS >5 for range of normal operation Material Machinability Maximize maximized Material Cost Minimize Within budget Material Weight Minimize Minimized for material Suspension Assembly and Measurement: For the FSAE suspension, the most important metrics are related to the geometry. For this reason, it is important that the suspension be fully assembled, so that these geometric metrics can be measured. Since the suspension is still in the process of bei ng assembled on the chassis, not all of the final values have been measured as of yet. One corner of the suspension can now be fully assembled for testing. A poster board will be fixed to the front side of the suspension box, so that instant centers and other geometric features can be marked. With the corner assembled, the relevant metrics will be measured. In order to meet the target value for anti-dive and anti-squat, the length of the switchable bushings is all that must be known. This was measured, using the distance between the tabs of the pickup point, the width of the rod end, and the variety of bushings that can be made.

9 Measuring the roll center position will be slightly more difficult. The suspension will be set in the neutral stance position, and the front-view instant center position constructed from the geometry and marked on the poster board. The line from the instant center to the tire contact patch will then be drawn, and assuming symmetry in the neutral stance, the roll center will be marked on the board. The suspension will then be set at the 0.5 jounce position, and the instant center reconstructed and marked. Since the roll center cannot be constructed from only one corner when the suspension is not symmetrically deflected, the instant center position of the one corner will be measured and compared the values calculated in the kinematic analysis to confirm the predicted movement. If the values do not match, further steps will be taken to assemble both sides of the suspension and physically measure the roll center movement. Measuring scrub radius will be a simple matter of measurement once the upright, clevis and wheels are assembled. The steering axis will be measurable from the clevises, and the distance between it and the wheel centerline at the contact patch will be easily constructed and compared to the simulated values. Unfortunately, this metric could not be significantly affected by any changes the team could make. The primary contributors to scrub radius are the front hubs, which were outside the scope of the project. This is the reason for this target value not meeting the goal value. The Camber is a critical measurement of the jig assembly. With the suspension set in the neutral stance, the camber angle will be measured by tracing the upright angle onto the poster board, and drawing and intersecting vertical line using a plumb. The angle between the two lines will then be measured. The suspension will next be set in the 0.5 jounce position, and the previous angle construction repeated. This new angle will be subtracted from the measured change in bottom a-arm angle to obtain the net camber change. The process will be repeated in the 0.5 rebound position. Kingpin angle was obtained by measuring the difference between the upright angle and the line between the two clevis centers, which constructs the steering axis. Since the upright clevises were all the same size, the kingpin angle as zero. To obtain caster angle, the side-view line between the top and bottom upright spherical joints, and then measuring the angle between this line and a vertical line constructed using a plumb. The caster angle will then be adjusted as necessary by changing the lengths of each arm in the a-arms until the desired angle is obtained. The number of tools required to adjust the suspension was easily measured from the suspension assembly, simply by noting how many tools were needed to adjust the key adjustability components of the suspension as it is assembled on the chassis.

10 The ease of adjustability metric will be a subjective measurement that is obtained by having each of the team members adjust the suspension, and describe their experience as either easy or not. This process will be repeated with members from the other FSAE teams, and receiving their feedback on whether the suspension was easy to adjust or not. Final Car Assembly: Some characteristics of the car simply cannot be measured before the suspension is assembled on the chassis, and the car is run through testing. These metrics include wheelbase and track. These values will have to be measured once the suspension and chassis are assembled and in running condition, at the end of winter session. Performance: All of the metrics and target values discussed above were developed with the maximum performance of the car as the goal. For the development of the project solution, it was important that the performance be broken down into these physical target values. However, once the car is assembled, it will be necessary to test whether the target values that were chosen through research lead to the desired performance of the car. Since the car will not be completed until the end of senior design, testing will take place over winter session and in the spring. For this reason, a test plan has been developed for the measurement of key performance characteristics. This plan will be left with the UD FSAE club, with instructions on how to measure the performance of the car, and what steps to take to fine tune and improve the performance. The key test procedures that will be developed are described below. Tire Temperature Analysis: Measuring the tire temperatures after time on the track allows the camber, ride rates, and roll rates, to be fine tuned. Using a tire pyrometer, the temperature of each tire will be measured in three places, 1 from the inside shoulder, 1 from the outside shoulder, and in the center of the tire. These three temperatures will be measured and recorded from each tire immediately after 5-10 track laps. The temperatures will then be analyzed, and the necessary tuning steps determined. A summarized diagnosis procedure is described in Table 3. Table 3: Tire Temperature Diagnosis Tire Temperatures Diagnosis Solution Individual Tire: Inside hotter than outside Slight difference, good Optimal temperature distribution Major difference, too much Add shims to top of upright. negative camber. Too much toe out Lessen toe out

11 Outside hotter than inside Not enough negative camber Remove shims from top of upright. Add shims to bottom of upright (only if absolutely necessary, toe will have to be readjusted.) Too much toe in Lessen toe in Sides hotter than middle Low Tire Pressure Increase Tire Pressure Middle hotter than sides High Tire Pressure Reduce Tire Pressure Between Tires: Rear Tires Hotter than front (oversteer) Front sway bar soft, rear stiff Stiffen front sway bar or loosen rear sway bar Rear tire pressure low, front high Increase rear tire pressure or Front Tires Hotter than rear (understeer) Front sway bar stiff, rear soft Front tire pressure low, rear high reduce front tire pressure Stiffen rear sway bar or loosen front sway bar Increase front tire pressure or reduce rear tire pressure Jounce, Roll, and Anti-squat and Anti-dive: These performance characteristics will be measured directly during performance testing of the car. Zip ties will be fit snugly around the shafts of the shocks, pressed up against the shock body when the car is in the neutral stance. The car will then be run through a series of maneuvers, and by measuring the distance that the zip tie has been pushed up the shaft, the wheel movement in jounce can be directly measured. The use of car-mounted video cameras has also been discussed for measurement of this and the other metrics described here. To measure anti-squat and anti-dive, the car will be run through maximum breaking and acceleration maneuvers. The car will be accelerated from a stop as fast as it is capable, and the deflection of the rear suspension will be measured afterwards from the zip tie movement. The accelerometer data from the data acquisition unit will be obtained, from which load transfer will be calculated, and compared to the calculated % anti-squat. A maximum breaking maneuver will be similarly performed to determine anti-dive. Measurement of body roll will be achieved by using the same zip tie method and running the car through maximum cornering maneuvers, replicating the skid pad event in competition. By measuring the amount of wheel movement using the zip tie method, the effective roll rate will be determined. Thus the contributions from the springs and sway bars to the roll stiffness can be measured and compared to the theoretical values.

12 The test plan outlined above will be submitted to the UD FSAE club as an independent document, to be kept in the club and used during the testing of the car, both by other club members and the members of the club in this senior design group. Transition: As our official responsibility for the project ends on the 17 th when the deliverables are presented to the sponsor, the importance of the transition to the UD FSAE club becomes very important. The test plan given above, and attached separately, will be a primary way that the team s understanding of the suspension will be passed on to the club. Additionally, participation of the team members within the club during the winter and spring will be important. Our involvement in the club will foster the transfer of knowledge about the suspension from the team to the club, preparing the junior club members for the potential senior design project in 2011. During the spring semester, team involvement in the club will help to test the durability of the suspension and tune it to its best performance. In the summer, at least a few of the team members will be seeking to travel to California for competition. This will be important for success in the static events, particularly the design event. To assist with this transitional process, and at the suggestion of the advisor, a list of tasks to be completed by the club is given here: Complete manufacturing of bushings. All of the different types of bushings have been manufactured, but not as many of each as are needed. These simple bushing simply need to be manufactured on an as-needed basis to complete the assembly of the suspension system. This includes primarily: o Bushing for the joints between the rockers and the pushrods, shocks, and sway bar linkages. o Bushings for the points where the shocks are mounted to the chassis. One of the front uprights needs to have its center hole milled out to match the larger diameters of the other front upright. The travel of the suspension needs to be confirmed (jounce and rebound), once the car is fully assembled. Due to rounded weld joints that were not accounted for in the solid model, the bottom edges of the top a-arms may need to be ground down at the upright side. Due to chassis/drivetrain complications over the location of the drive chain, the frame member that the rear shocks are mounted to may need to be moved. The shock mounts and possibly the rear rocker mounts may need to be altered to fit the new frame members. The full welding of all the tacked suspension/frame joints needs to be completed. For suspension pickup points on chassis, C-shaped side gussets need to be cut and welded to the pickup tabs.

13 All aluminum suspension components should be anodized and all steel components powder-coated. Voids in the a-arms created by the parallel gusset plates need to be filled with foam or capped to prevent debris filling the voids. Nut faces should be added to the pushrods, tie rods, and sway bar linkages, to make adjustment easier. This can be achieved by drilling out the center of a nut, sliding it over the tube of the rod, and welding it in place. When shims are manufactured for the uprights, they need to have tabs on the top or sides to make addition and removal of tabs easier. For the adjustable anti-dive and anti-squat pickup points, multiple bushing sets should be made, to allow adjustment of the rod end up and down in ¼ increments. Push/Pull testing should be performed on duplicate components of the adjustable anti pickup points, to validate the strength of the points. Conclusion: Team FSAE Suspension is glad to have reached completion of this project. At the beginning of the semester, we knew close to nothing. Now, as we finish up the project, we are confident in our knowledge of formula-style suspension, in theory and with some experience in practice. The project began with an extensive amount of research, to educate the team and to develop key metrics and target values. The sponsor and advisor were invaluable in this process, providing a large amount of in-depth information on suspension and its workings. Moving forward from metrics and target values, concepts were proposed and analyzed, until the top ideas had been sorted out and developed. From there, the concepts were developed into detailed designs, with the end result of component and assembly drawings in preparation for manufacturing. Phase 4 was almost entirely consumed with manufacturing and fabrication, transforming the CAD models and drawings into real parts. This required a lot of growth among the team members in machining and welding ability. The team considers the project a success, as all the parts have been manufactured, and by the 17 th, it will all be finally attached to the chassis. Testing and tuning in the winter and spring will confirm the performance characteristics of the car, and the team is confident that the results will be positive and as expected. In conclusion, Team FSAE Suspension has enjoyed the opportunity to participate in the FSAE sections of senior design. The project assigned us was very in-depth and involved, and possibly too much to be realistic for one semester, especially considering the amount of manufacturing involved. However the team was able to complete the project on schedule, with a design that meets the target values and goals set forward at the beginning of the project, while also staying within the constraints imposed on us. Many thanks are due to Dr. Steven Timmins, our advisor, who provided oversight of the project, along with huge amounts of help through suggestions

and discussions of concepts and theory. We are also indebted to Steve Beard and Al Lance, who patiently helped us with manufacturing, and worked late to provide us with evening hours, respectively. Moving forward, the team is confident in the future success of the suspension, and looks forward to staying involved over the winter and spring, and proceeding to a successful trip to competition at the beginning of the summer. 14

15 APPENDIX UD FSAE Suspension Testing and Tuning Instruction Manual Senior Design Team FSAE Suspension December 10, 2010

16 Introduction: Over the course of the senior design class of Fall 2010, the FSAE Suspension team developed the suspension system used on the 2010 UDFSAE car. Due to the nature of the course, and the depth of the project, fabrication of all of the car s system was finished only days before the deliverables deadline at the end of the semester. For this reason, the team was not able to complete any kind of performance evaluation of the suspension system. This manual outlines a set of test procedures intended to evaluated the performance characteristics of the suspension, and provide guidelines for tuning the suspension to achieve maximum performance. Purpose: All of the geometry-related metrics and target values developed during senior design were intended to maximize performance. Appropriate camber, ride and roll rates, and anti -dive and anti-squat all serve to improve the performance of the car on the track. Theoretical target values were developed during the semester, all with the goal of giving the car the best possible performance. However, to actually achieve the best performance, it is necessary to tune the adjustable parts of the car based on the results of testing and driving. The purpose of this manual is to give some direction to the tuning of the car, making it easier to achieve optimal performance. Equipment: 1. Tire Pyrometer 2. Zip Ties 3. Video Camera preferably car-mounted 4. 1 9/16 wrench 5. 2 7/16 wrenches Procedures: Ride Height: To adjust the ride height of the car: 1. Using a 9/16 wrench, loosen the jam nuts on either end of the push rods.

17 2. Turn the pushrods to either lengthen the pushrods to raise the ride height, or shorten the pushrods to lower the ride height. 3. Retighten the jam nuts. 4. If the incorrect ride height is caused by the shocks being excessively compressed, or not compressed enough, adjust the preload on the shocks as necessary. 5. Make sure to keep the ride height the same from side to side. Front to back can be different as desired. Static Camber: To adjust the static camber of the car at ride height: Toe: 1. Using the two 7/16 wrenches, loosen the two bolts on the top a-arm clevis at the top of the upright. 2. Insert shims as necessary to add to the camber, or remove them to remove positive camber. 3. If more negative camber is desired than be achieved by removing all the shims, loosen the bolts at the bottom of the upright and add shims. 4. Retighten all the bolts. To adjust the toe of the wheels 1. Using the 9/16 wrench, loosen the jam nuts on either end of the tie rod. 2. Turn the tie rod as necessary to lengthen or shorten it, and toe out or in. 3. Retighten the jam nuts. Tire Temperature Analysis: Measuring the tire temperatures after time on the track allows the camber, ride rates, and roll rates, to be fine tuned. To perform a tire temperature analysis, complete the following steps: 1. Bring the tires up to operating temperature by driving the car. Perform several laps of the test course. 2. Immediately after performing the tire warm-up, use the tire pyrometer to measure the tire temperature as shown in Figure 1. Measure in three places: 1 from the inside shoulder of the tire, 1 from the outside shoulder, and in the center of the tire.

18 3. Record the temperatures and repeat for all four tires. 4. Using Table 1, diagnose the tire temperatures and perform the solutions as directed. Table 1: Tire Temperature Diagnosis Tire Temperatures Diagnosis Solution Individual Tire: Inside hotter than outside Slight difference, good Optimal temperature distribution Major difference, too much Add shims to top of upright. negative camber. Too much toe out Lessen toe out Outside hotter than inside Not enough negative camber Remove shims from top of upright. Add shims to bottom of upright (only if absolutely necessary, toe will have to be readjusted.) Lessen toe in Too much toe in Sides hotter than middle Low Tire Pressure Increase Tire Pressure Middle hotter than sides High Tire Pressure Reduce Tire Pressure Between Tires: Rear Tires Hotter than front (oversteer) Front Tires Hotter than rear (understeer) Front sway bar soft, rear stiff Rear tire pressure low, front high Front sway bar stiff, rear soft Front tire pressure low, rear high Stiffen front sway bar or loosen rear sway bar Increase rear tire pressure or reduce front tire pressure Stiffen rear sway bar or loosen front sway bar Increase front tire pressure or reduce rear tire pressure Jounce, Anti-Dive, and Anti-Squat: The stiffness of the suspension and its anti-dive and anti-squat performance are closely linked. The amount of dive experienced during braking, and the squat during acceleration, are direct results of both the spring stiffness and the geometry of anti -dive and anti-squat. To adjust the jounce and anti features, follow the steps below: 1. Strap small zip ties tightly around the shafts of the shocks, and slide them up the shock until they are sitting against the body of the shock. 2. Run the car through a straight line accelerating and braking test, accelerating as fast as possible and locking up the wheels during braking. It is important to not make any sharp turns with the car during this test.

19 3. Measure how far the zip ties were pushed down the shaft on all four shocks and record the values. 4. Generally, the stiffer the suspension, the better. If it is decided that the car is diving excessively during braking or squatting during acceleration, replace the bushings on the adjustable pickup points at each cornering, lowering the point. 5. Rerun the test, re-measuring the anti features and adjusting again as necessary. 6. If the dive and squat are still excessive, or cornering performance is negatively affected by increasing the anti features, stiffen the cars preload or replace the springs with stiffer ones. 7. All of the solutions given in this section must be performed in conjunction with other tests outlined in this manual. For example, stiffening the springs or adding preload will also affect the roll of the car in corners, so the performance of the car in this respect should be retested after any stiffness changes are made to the car in anti-tuning. Body Roll: The body roll is one of the most important performance characteristics of the car. The amount of roll the car undergoes during cornering significantly affects the camber changes of the wheel, and cornering force the car can generate, and other features. As with other suspension characteristics, in general stiffer is better. Test the cars roll using the following steps: 1. It is difficult to physically measure the roll of the car. It must be judged by the driver s feel, and visual evidence if video can be taken of the car in testing. Run the car through a skidpad test as will be run in competition, a 30 ft radius circle. Based on the driver s input, observation, and video footage if possible, determine if the car is rolling excessively. If so, unbolt the sway bar linkage from the sway bar arm and re-bolt it to a hole closer to the base of the arm. 2. Rerun the test. If the roll is still excessive and the sway bar linkage is attached to the shortest bolt hole, add pre-load to the shocks or replace the springs with stiffer ones. However, make considerations of the affects this will have on other performance characteristics before making changes to the spring stiffness. 3. If the role is still excessive, and/or the springs cannot be changed due to other performance factors, replace the sway bar with a stiffer one. 4. In the case that the roll is too stiff, perform the opposite of the steps above.

20 Notes: All of the performance characteristics discussed in this manual are inter-related, and in most cases adjusting one will affect the others. This must be kept in mind when adjustments are made. No large changes should be made to any one feature without then testing the car again to determine if other performance areas have been altered. For questions regarding the 2010-2011 suspension, its design and tuning, contact: o Seth Beckley sethbeck@udel.edu