Background Information and Research - University of Idaho



MemoDate:December 15, 2008To:Don Blackketter, Edwin Odom, and Steven BeyerleinFrom: H-FSAE teamSubject: HyRollers End of Semester ReportAttached is a report outlining the HyRollers’ research, conclusions and accomplishments for the summer and fall semesters of 2008. Great care has been taken to accurately document the progress of the first phase of the University of Idaho’s development of a completion winning hybrid design for the Society of Automotive Engineers Hybrid Formula race car competition (H-FSAE) in 2010. It was our intention to lay a solid foundation on which subsequent teams can build on. Based on the information gathered and reviewed over the summer semester, we recommended the complex hybrid configuration. This design afforded the most versatility at the lightest weight while playing to the strengths of the University of Idaho. However, due the high degree of complexity of this system, a scale model was conceived, designed and built to verify its feasibility and the suitability of the use of DC brushless motors. The two motor test bed has been completed and is ready for additional rounds of testing to determine the final drive ratios needed for the full scale planetary system. Additionally this platform will serve as the testing grounds for the forthcoming electrical engineering team’s power management hardware and software. The data collected to date has shown that the very favorable power density of the DC brushless motor may indeed be a great fit for this application. This test bed will not only serve as a tool for designing a world class hybrid, it will also function as an impressive show piece for the University. We have all greatly appreciated the opportunity to work on this challenging and rewarding design problem, and look forward with much anticipation to the 2010 H-FSAE competition. Thank you for your support throughout our capstone experience and please feel free to contact any of us if you have further questions or concerns.HyRollers Hybrid Driveline Development for H-FSAE CompetitionFinal Design ReportPrepared for:Dr. Blackketter, Dr. Odom, and Dr. BeyerleinPrepared by:John Whitchurch, HyRollers team memberChris Gullickson, HyRollers team memberTroy Vandenbark, HyRollers team memberJedidiah Bartlett, HyRollers team memberDecember 15, 2008Executive Summary“Powertrain Recommendations for the U of I Entry into the H-FSAE 2010 Competition”Beginning in the summer of 2008, the HyRollers where given the vague task of understanding and becoming experts with basic hybrid vehicle design concepts. This transformed quickly into a project of intense research striving toward our main goal, performance. ?The team discovered that there is no accurate way to compare various configurations to one-another without doing a preliminary design and component selection for each configuration to be evaluated. In order to forge ahead the HyRollers focused on the configurations that played to the strengths of the University of Idaho Mechanical Engineering Department. The two designs considered were the Parallel system and the Complex system.Because of the great success the 2007University of Idaho SAE formula car team had with their custom made planetary differential, Dr. Odom and the HyRollers felt confident in attempting the same feat in a complex hybrid using a similar planetary gear box. The basic design is to approximate the drivetrain presented in Jason S. Sagen’s thesis entitled: HYBRID-ELECTRIC PLANETARY DRIVETRAIN DESIGN FOR AUTOMOTIVE APPLICATIONS.? This uses a planetary input differentiating system in order to allow either the engine, or an electric motor, or both to drive the output of the planetary system.? This design will also use “torquer” motors/generators on the output to add to the torque supplied by the engine and to recapture energy from regenerative braking.Before a full scale version was to be attempted, it was decided to vet the complex system by building a scale model version of the design proposed in the Jason’s thesis. This task was broken into several phases. Step 1: Relating the Test-Platform to the Real Hybrid Step 2: Finishing the Dynamometer Step 3: Modifying the Planetary Gear System left by Jason Sagen Step 4: Designing and Building the Single Motor Test Platform Step 5: Testing on the Single Motor Test Platform Step 6: Design and Build of the full Dual-Input Test Platform This effort has produced a scale model, dual input, planetary drive that can be used as both a test-platform for future development, and a teaching aid for future teams. The HyRollers have spent a great deal of time and effort documenting their work to make it easier for future teams to reference and for them effectively pick up where they left off.Table of Contents TOC \o "1-3" \h \z \u 1.Background Information and Research PAGEREF _Toc217136920 \h 11.1H-FSAE Competition PAGEREF _Toc217136921 \h 11.2University of Idaho H-FSAE Goal: PAGEREF _Toc217136922 \h 11.3Hybrid Technology and Development PAGEREF _Toc217136923 \h 12.Problem Definition PAGEREF _Toc217136924 \h 32.1Opportunity Statement: PAGEREF _Toc217136925 \h 3To build and test a scale model drive train for proof of concept. ?To use scale model testing to make a recommendation for future Hybrid FSAE Teams. PAGEREF _Toc217136926 \h 32.2Design Specifications PAGEREF _Toc217136927 \h 32.3Team Deliverables PAGEREF _Toc217136928 \h 33.Project Plan PAGEREF _Toc217136929 \h 43.1Major Milestones in the Project PAGEREF _Toc217136930 \h 43.2Team Responsibilities PAGEREF _Toc217136931 \h 44.Concepts Considered PAGEREF _Toc217136932 \h 54.1Series Hybrid Systems PAGEREF _Toc217136933 \h 54.2Parallel Hybrid Systems PAGEREF _Toc217136934 \h 64.3Complex Hybrid Systems PAGEREF _Toc217136935 \h 65.Concept Selection PAGEREF _Toc217136936 \h 85.1Initial Method of Selection PAGEREF _Toc217136937 \h 85.2Final Method of Selection PAGEREF _Toc217136938 \h 85.3Concept Proposed PAGEREF _Toc217136939 \h plex Configuration Miniature Test Platform PAGEREF _Toc217136940 \h 106.1Detailed Design Outline PAGEREF _Toc217136941 \h 106.2Summary PAGEREF _Toc217136949 \h 167.Future Work PAGEREF _Toc217136950 \h 167.1Test-bed specific work (upgrades and wrap-up) PAGEREF _Toc217136951 \h 167.2Future Testing that should be performed PAGEREF _Toc217136954 \h 177.3Further Development of Hybrid system using the Scale Test-Stand PAGEREF _Toc217136957 \h 188.Work Cited PAGEREF _Toc217136962 \h 18 Table of Figures TOC \h \z \c "Figure" Figure 1: H-FSAE Competition Points Breakdown PAGEREF _Toc217146037 \h 2Figure 2: Outline Diagram of the Design Path PAGEREF _Toc217146038 \h 2Figure 3: House of Quality Illustrating Important Design Considerations PAGEREF _Toc217146039 \h 4Figure 4: Gantt Chart showing design milestones PAGEREF _Toc217146040 \h 5Figure 5: Energy Flow for a Series Hybrid PAGEREF _Toc217146041 \h 6Figure 6: Energy Flow for a Parallel Hybrid PAGEREF _Toc217146042 \h 7Figure 7: Basic Energy Flow for the Complex Hybrid PAGEREF _Toc217146043 \h 8Figure 8: Complex Configuration Morphological Chart PAGEREF _Toc217146044 \h 9Figure 9: Energy Flow for Complex Configuration Proposed PAGEREF _Toc217146045 \h 10Figure 10: Planetary Configuration Modeled PAGEREF _Toc217146046 \h 12Figure 11: Jason Sagen’s Proposed Planetary Configuration PAGEREF _Toc217146047 \h 13Figure 12: Single Motor Testing Schematic PAGEREF _Toc217146048 \h 14Figure 13: Calibration and Data-Collection PAGEREF _Toc217146049 \h 15Figure 14: Data Points Collected when doing the Linear Calibration PAGEREF _Toc217146050 \h 16Figure 15: Data Reduction, Isolating Readable Data PAGEREF _Toc217146051 \h 17Figure 16: Torque Curve for A40-14S Hacker Brushless DC Motor PAGEREF _Toc217146052 \h 18Background Information and ResearchH-FSAE CompetitionThe Society of Automotive Engineers (SAE) has recently endorsed the collegiate Hybrid Formula Race Car (H-FSAE) competition. It is the intent of the University Of Idaho College Of Engineering to have a completed car for the 2010 competition. The summer-fall, 2008 inaugural H-FSAE team (designated HyRollers) has taken on the responsibility to select the hybrid platform which provides the greatest opportunity for success in 2010. This decision was required to allow the University of Idaho to allocate their full resources to the multitude to subsequent design considerations entailed. There was simply not enough time or manpower available to fully investigate each of the layouts. Therefore, further research was done to qualitatively compare the three industry-standard configurations based on the idealistic performance of each. The debate over the merits of each system rages among professional engineers throughout the automotive industry. Well formulated arguments can be made to support any of them. University of Idaho H-FSAE Goal:This project supports NIATT Goal 3 Strategy 3.3 to increase the number of students in the transportation workforce by a viable University of Idaho entry in the SAE sponsored hybrid FSAE competition scheduled for 2010.? The primary focus of year 1 is demonstration of a robust hybrid power plant design consisting of an internal combustion engine, hybrid transmission design, and motor selection.? The primary focus of year 2 is system integration of the hybrid power plant into a competition platform.? In the process the University will engage a broad population of undergraduate students in implementation of sustainable transportation technology and expand local design analysis as well as advanced manufacturing infrastructure. Throughout the duration of the FSAE Formula Car project at the University of Idaho we have focused on the demonstration of student designed and manufactured subsystems in coordination with a GSX-R600cc motorcycle engine. In the past, the spirit of the rules revolved around the design and construction of a high performance gasoline race vehicle with prohibition of any modified drive train to include electric motors.? The proposed project is in compliance with the global trend to research environmental sustainable transportation systems as well as aligning with the College of Engineering’s “Go Green” motto.?Hybrid Technology and DevelopmentBy convention, a hybrid is specifically designed with efficiency as the utmost concern. However, for this application, performance is paramount. Hypothetical race performance characteristics of each of the three configurations needed to be compared with a special emphasis on high electrical/IC power ratio, electric only acceleration, unrestricted acceleration, and endurance (13). This need stems from the H-FSAE competition rules and regulations. The competition is made up of several categories which are independently scored on a 1000 point scale as shown in table 1 below (14).Static EventsPointsPresentation100Engineering Design200Dynamic EventsAcceleration (Electric only)75Acceleration (Unrestricted)75Autocross150Efficiency and Endurance400Total points1000Figure SEQ Figure \* ARABIC 1: H-FSAE Competition Points BreakdownA comprehensive analysis of options available, selection of the most promising option, and the tools to move that option forward will give future teams that build on the HyRollers work the ability to go to competition with a winning design that will establish the University of Idaho as a leader in hybrid technology research and development. The HyRollers team has built that foundation and developed a roadmap to bring an award-winning design to fruition. This road-map can be seen below in Figure 2.Figure 2: Outline Diagram of the Design PathProblem Definition Opportunity Statement:To build and test a scale model drive train for proof of concept. ?To use scale model testing to make a recommendation for future Hybrid FSAE Teams.Design SpecificationsAs a team it was decided that we needed an innovative design that could be competitive in both the design portion of the competition as well as in the acceleration and endurance runs. For this we feel that we have a highly innovative design that will garner good reviews in the design portion. For the dynamic events, we feel that we need an emphasis on lightweight, rapid acceleration and a moderate top speed. The H-FSAE competition is trending towards more aggressive track courses with many hard corners and short straight-aways. Such courses will favor hybrids with effective regenerative braking and strong acceleration and deceleration capabilities. To take full advantage of these considerations, our design will combine a lightweight model with snappy acceleration and moderate top speed while maximizing regenerative braking. The design will run the I.C. engine at a constant rpm at its maximum power output so the car will have the highest available power to win the endurance race portion. A core strategy for the design of our hybrid is to burn more gas than any of our competitors. This philosophy runs counter to traditional hybrid thinking because traditional hybrids are designed with maximizing efficiency as the main goal. Our application of hybridization is to get to boost performance not efficiency. By burning more gas than any other team we effectively have more total energy at our disposal than any other team. Until the rules of this competition are changed to give efficiency a larger share of points it is essential to keep performance a priority in all aspects of design. If this design strategy can be properly implemented then the University of Idaho should have a highly competitive vehicle at its first competition. These Goals stated here can be summarized as below:Develop an innovative Hybrid drivetrainDeliver the maximum amount of power possible from the IC engine (Burn a lot of gas)Ensure that this design is lightweight and maneuverableDesign it to be the foundation for an award-winning car in competitionThe Team also developed what is known as a House of Quality in order to fully describe all of the design specifications. This House of Quality is shown below in Figure 3.Figure 3: House of Quality Illustrating Important Design ConsiderationsTeam DeliverablesBy the end of the fall 2008 Semester, the HyRollers team was able to deliver the following items in accordance with achieving the above goals.T-drive database organizationResearch and design documentationThe design and construction of a test bed for single and dual motor inputsExperimental procedures for both test bedsHacker Brushless motor torque data to be used to interpolate the large/expensive Hacker motor torque characteristicsProject Plan Major Milestones in the ProjectFigure 4: Gantt Chart showing design milestones-580390467995The major decisions and work that was done on the project can be seen in the following Gantt chart:Team ResponsibilitiesThe Team divided up the responsibilities for the project completion into roles and decided as a group who should fill these roles. The specific duties were divvied out with consideration for those that had experience or talents that would suit that particular role. If there was no one that had experience in one of these specific areas, then someone who wanted to have experience in that area was appointed. The whole team was responsible for all aspects of the project and it was agreed that each would help the other so that we could all be involved in the whole process. Each of these roles then became the guideline for whom to assign specific tasks to, and when one person was over-worked, the others would take some of the work off of their shoulders. The general team roles were as follows:Database Manager and Shop Liaison – Chris GullicksonTeam Webmaster and Draftsman – Troy VandenbarkPurchasing Agent/Puppet Master – John WhitchurchTeam Leader – Jedidiah BartlettConcepts ConsideredThere are three major types of hybrids on the market; these were investigated over the course of the summer semester. These hybrids were evaluated based on the pre-defined goals that were laid out during the Problem Definition segment of the design process. A major design consideration in all of these models is that the internal combustion engine remains the only energy input to the system. The hybrid system then, cannot amplify or provide additional power, but only redistribute the power being produced by the engine by storing it in a battery when the engine has extra power to spare, and delivering it to the wheels when power requirements exceed the output of the engine. This allows the engine to be maintained in a range closer to either it’s most efficient range, or in our case, its maximum maintainable output range.The configurations as presented here are gross oversimplifications, but provide the reader a sense of how these different layouts work.Series Hybrid SystemsSeries hybrid systems (2) are the simplest type of hybrid. A series system generally consists of some kind of internal combustion engine (ICE) driving a generator. Mechanical energy is converted to electrical energy at the generator. This electrical energy is stored in some type of energy storage device, usually a battery pack or ultra capacitors bank. The electrical energy stored within the battery pack powers an electric motor. The electric motor powers the vehicle, usually through a typical automatic or continuously variable transmission. In some applications no transmission is used and the motor (or motors) drive the wheels directly. The energy-flow for this system can be seen in Figure 5Figure SEQ Figure \* ARABIC 5: Energy Flow for a Series HybridAn example of a serial hybrid is a locomotive. The large motors required to drive the wheels of a locomotive are acceptable due to immense size and weight of the locomotive itself. For a smaller vehicle, the size of the required motors would not be practical, so smaller, high speed, low torque motors are used. The torque boost and speed reduction from the high speed motors is accomplished via the gearing in a transmission.Series configurations have found favor among large-vehicle fleets due not only to their simplicity, but due to the operational nature of fleet vehicles, such as buses and locomotives. These large fleet vehicles generally have very predictable routes, allowing drive train components to be tailored specifically to the requirements of the route. Passenger vehicles generally have a much more chaotic driving cycle, making it difficult to design a suitable drive train (16). General Motors was one of the first companies to experiment with a series hybrid passenger car. In 1968 General Motors built and tested a hybridized Opel Kadett. It was powered by an 8hp Sterling engine. Even though the car weighed almost 1000 lbs more than the original Opel, it still managed to accelerate from 0 to 30mph in about 10 seconds, had a top speed of 55mph and achieved 30-40 miles per gallon (15). The design’s major drawback is that only one of the rotating components actually has the potential to drive the wheels. This means that were there would normally be only an IC engine, a generator and electric motor are added to the weight without increasing the average amount of energy delivered to the wheels. If the final drive motor is more powerful than the IC engine, it can be made to deliver higher burst power, but it can only maintain the output of the engine (minus the system losses) as it is the only input to the system.Parallel Hybrid SystemsThe classic parallel configuration shown in Figure 6 combines many of the same components from a series configuration. The gasoline engine is coupled directly to the transmission. A battery pack (or ultra capacitor bank) provides electrical energy to an electric motor. The electric motor also provides input to the transmission. The transmission output provides torque to the wheels. Generally the engine and electric motor are linked to the transmission input via clutches. This allows for both engine and motor to be engaged or disengaged as needed. The ability to disengage both power sources independently means that the vehicle can be propelled entirely by the engine, entirely by the electric motor, or from a combination of both. Since a parallel hybrid configuration has two independent means of providing torque to the wheels, neither source needs to be as large as for a comparable series configuration (1). Also, since the amount of power each source provides can be varied as needed, the parallel configurations lends itself well to applications where power demand is constantly changing, such as stop and go driving. Unfortunately, the classic parallel configuration suffers from the same major drawback of an entirely electric car, that is, the need to recharge the battery pack externally. In order to overcome this, most of these systems use a small alternator to recharge the battery pack slowly from the IC engine, making it ideal as a launch-assist system, but may not be feasible for the cyclic acceleration demands of racing.Figure SEQ Figure \* ARABIC 6: Energy Flow for a Parallel HybridComplex Hybrid SystemsThe most recent hybrid configurations have trickled down from the series and parallel hybrids and depending upon the source, have one of several names from complex hybrids to power assist or dual mode hybrids. Whatever their name, these hybrids are able to combine characteristics of the series and the parallel configurations. A complex hybrid replaces both the electric motor and electric generator in the series configuration with a pair of electric motor/generators (M/G 1 and M/G 2), as seen in Figure 7.Figure SEQ Figure \* ARABIC 7: Basic Energy Flow for the Complex HybridBecause each motor/generator has the potential to either convert electrical energy to mechanical energy, or to convert mechanical energy to electrical energy, power is allowed to flow either direction through them. Electrical energy can also flow directly from M/G 1 to M/G 2. The complex hybrid allows for multiple operating conditions. Mechanical energy can flow from the engine to the wheels directly, or it can flow into M/G 1 and be converted into electrical energy to charge the battery. Electrical energy stored in the battery can flow to M/G 2 where it can either power the vehicle independently, or help supplement the power coming directly from the engine. Power from the battery can also flow through M/G 1 to the engine, which allows M/G 1 to replace the starter motor found on a traditional ICE. Mechanical power can flow from the wheels, through M/G 2 and help charge the battery; this is known as regenerative braking and is one the most important ways hybrids increase fuel economy. Obviously a complex hybrid can also run in any combination of the above conditions (12).The complex configuration has all of the benefits of the parallel configuration, along with most of its drawbacks. As discussed above, the control of the components of the complex hybrid is difficult. This section has described the mechanical design of the complex hybrid, but the control system logic of the Toyota Prius, a complex hybrid, is explained in depth in “Investigation and Simulation of the Planetary Combination Hybrid Electric Vehicle” (15). The complex hybrid configuration requires that power from both of the electric motor/generators, the engine, and the wheels all be mechanically linked. There are several ways to achieve this, including the use of a planetary gear set. A planetary gear set consists of a sun gear surrounded by three or more planet gears. The sun and planet gears are enclosed by a ring gear (12). The planetary gearbox provides three separate Input/output shafts, each with a different drive ratio. The ability to adjust this relative mechanical advantage of each of the I/O shafts permits the marriage of a relatively small, high speed electrical motor with a much larger and higher speed IC engine to the low speed high torque requirements of the car drive line.Concept SelectionInitial Method of Selection The decision for selecting which design to pursue was a difficult one. The team tried many different methods to develop a mathematical model where they could “race” a proposed vehicle around a simulated track and by changing the configuration substantially (See Appendix D). The team also acquired an educational license to PSAT (Appendix E) an industry standard plug-in for Simulink that allows a graphical, model based approach, for simulating a given hybrid system and monitoring its power-flow characteristics during simulated road loading conditions.However, in order to use either of these systems, or a combination of the two, a preliminary design must be done for each configuration to be tested. This is well beyond the scope of a one year long design project. With preliminary calculations, the team managed to prove that any of the three configurations could be made to perform very well in the competition. This conclusion was backed up by the fact that there were several different configurations that performed well in the 2007 H-FSAE competition. (Appendix B)Final Method of SelectionThe final decision was made by process of elimination rather than any analytical method. The available methods proved to be too cumbersome and did not lend themselves to being a design tool that could help to define which components and configurations to use. However, these software packages will be very useful in the evaluation of a finished design. It was decided that the series version would not be used. This was because it could not be made to be light and the team determined that having a “quick” car was much more advantageous than having a powerful car. The basic parallel system was not desired because it is essentially two drivelines in one car and also cannot be made to be a light vehicle. These claims were again substantiated by analysis of previous teams and the fact that all of them were substantially heavier than the larger FSAE cars that contained larger engines. Figure 8 below shows that the complex hybrid could be constructed at about the same weight as the FSAE cars (~500 lbs).Complex Hybrid?Weight (lbs)Cost ($)Engine??Yamaha YZ 25068 lb$1500Kawasaki Ninja 25070 lb?Generators/Speeder??Hacker8 lb$1000Batteries??Lithium Ion0.2976 lb?Maxwell Ultra caps (50)40.418 lbunknownAdvanced Lead Acid100 lb?Elect Motor??Torque Motor40 lb$950Total WeightTotal Cost513.7 lb?$3450 unknownWeight of Frame and Yamaha 250 alone (lbs) – 425lbsFigure SEQ Figure \* ARABIC 8: Complex Configuration Morphological ChartConcept ProposedThe version of the complex drivetrain chosen by Jason Sagen in his Master’s thesis was based on Toyota’s Power Split Device, a planetary gearbox that allowed two inputs to the final dive train (used in the Synergy drive). For the internal combustion engine to be used in the final car, the Team chose the YZ250F due to its very competitive torque characteristics and the fact that we already have a solid model for this engine supplied by a previous senior design team. As for the speeder, the Hacker DC brushless motor appeared to have the highest power to weight ratio of the various motors we researched. The Maxwell Ultra Capacitors were by far the best of the three choices for the accumulator because of their superior energy density over the other ultra caps. However, they are much more expensive than the other choices. The team will leave the choice of the M/G 2 and 3 in the figure below to future teams as it is not obvious whether a Torque or pancake motor should be used because of their low-speed, high torque characteristics that do not require gearboxes, or additional Hacker Brushless DC motors because of their high power-weight ratios but will have to be geared down. Figure 9 shows the basic layout of the complex configuration as well as the way the planetary geartrain will integrate the YZ 250F (IC Engine), the speeder, and the torque motors to the drive wheels.Figure SEQ Figure \* ARABIC 9: Energy Flow for Complex Configuration ProposedThis configuration eliminates the need for a conventional transmission and replaces it with the planetary differential and the speeder motor. The capacitors are used rather than a battery pack because it was determined that the average power output can be provided directly from the engine, and that only short bursts of torque would be required for acceleration. Additionally, short bursts of charging from hard breaking into corners will be absorbed by the capacitor bank. Ultra capacitors are much better suited to this application than batteries because of their extremely high power density allowing for very high rates of charge and discharge. The reduction in the planetary differential and the final drive differential should be sized so that the internal combustion engine does not require any speeder input to maintain the average speed on the track. During slower portions of the track the speeder will act as a generator and slow the final output down without diminishing the torque. During high speed portions of the track, the speeder will add power at the same torque as the engine to over-speed the final drive and allow for a higher top speed. It is difficult to control this configuration because power can flow in almost any direction, and because of this it was decided that a scale-model be built to further develop the concept. The team believes this design will be race worthy, as Toyota has used it in various racing platforms (19).Complex Configuration Miniature Test PlatformDetailed Design OutlineThe design of the two motor test platform took its inspiration from the current Synergy Drive on the Toyota Prius. This driveline is complex and requires an advanced control system to be able to properly balance the various power supplies. In order to do that, a scale test-platform was built to accommodate two power inputs and combine them through a planetary gear-drive in order to allow one or the other, or both to provide output. This output runs into a small water-brake dynamometer that measures the amount of torque applied. Section 6.1.1 describes how the design was approached, and the sections following that describe, in steps, how the design was carried out. The steps outlined in the following sequence did not need to occur in exact chronological order, but they did help to divide the project into easily explainable sections.Outline of the Final Drivetrain Configuration to be modeled on the test-platformThe first thing that had to be done by the team once a course of action was selected, was to lay out the design parameters for this test-platform. The test-platform should be made to approximate the actual hybrid driveline so that whatever was modeled on the platform would scale properly to the final effort. The most important things to consider were; which of the inputs should be connected to which shaft on the planetary gear system, and how much of the system should be modeled together?The team first decided that the torque motors that are on the front wheels (shown in Figure 9 in section 5.3) did not need to be included in this first revision of the planetary model, but should be considered in the design so that their effects could be added later.Second, which of the three remaining components should connect to the planetary system in which places? This was decided by looking at the range of speeds that the various components operated over. The IC engine runs between 1000 and 18 000 RPM. This is a very large range, and also has the highest top-end of any of the rotating components. Therefore it was decided that it should be connected to the sun gear, which is the smallest effective gear and allows reduction when power goes out to the speeder, or out to the wheels, or a combination of both. The second highest rotating component is the speeder motor, which can spin between 300 and 8000 RPM (estimated). This was connected to the middle sized effective gear, the ring gear. Finally, the final drive on the wheels needs to turn between 0 and 4000 RPM based on the wheel diameter and the maximum speed desirable. This was connected to the carrier, which is the largest effective gear in the system. Figure 10 below illustrates the configuration chosen.Ring Gear – Speeder (Motor Generator)Carrier for Planetary Gears – Out to Final DriveSun Gear – IC EngineFigure SEQ Figure \* ARABIC 10: Planetary Configuration ModeledStep 1: Relating the Test-Platform to the Real HybridOne of the main advantages to this model is that the speeder motor/generator can have its ratio changed to accommodate different motors or optimize the design very easily. On the Test-model, both the IC engine and the speeder motor can have their ratios changed by purchasing different pulleys and belts. But on the actual drivetrain, only the final drive to the road can be changed easily. However, if the future teams that work on packaging make this a priority, the speeder can be connected to the outside of the ring gear via another gear or by using a belt and pulley. This would allow the team to optimize the ratio of the speeder motor to a specific track. The lower the ratio of the speeder, the higher torques will be transmitted to the wheels, at the expense of speed range. There will also have to be at least one clutch and internal brake on this system. The brake would stop the ring gear and motor-generator, creating a direct connection between the engine and the road. This would be used when the system is near “equilibrium” and the speeder is not substantially changing the desired output speed. In this condition, “locking” the electric motor in place requires a substantial amount of current, locking up that shaft with a brake is more efficient and will allow smoother transitions from under-speed condition where the speeder is absorbing energy from the motor and spinning backwards (charging the batteries) and over-speed condition where the speeder adds speed to the wheels above the speed the motor alone can give. The Clutch would be used between the engine and the planetary so that the engine could sit and idle without being connected, and this clutch could be “dropped” at the beginning of a race like one would normally do to provide a high starting torque using the rotational inertia of a wound-up engine. A similar thing can be done to the speeder with this clutch. Because having the engine clutched out from the planetary system, any input from the electric speeder would cause the other half of the clutch to free-wheel rather than turning the final drive. This spinning flywheel, along with the clutch on the engine side could both be locked up together by dropping that clutch. And finally, it may be necessary to utilize this flywheel effect when starting the motor if the gear-ratios do not allow the electric speeder enough torque to turn over the engine. If this flywheel effect is necessary to start the engine, it should not count as an “energy storage device” and should therefore be allowed by the judges at competition (flywheels as energy storage devices are not allowed) Future teams may require more clutches and breaks than this, but all of this will be left to them as they continue to flesh out the design we have laid out. The rest of this paper will, instead, focus on the tool the team has developed to be used by other teams.Future teams may also want to include a flywheel onto the dynamometer shaft to give the model inertia so that regenerative braking systems can be tested, and one or two “torquer” motors could be added to the same shaft that will simulate the two additional motors on the front wheels so that power-balancing between all of the rotating components can be checked.Step 2: Finishing the DynamometerThe first thing that the team had to design is the data-aquisiton for the waterbrake dyno. This had been left undone by the builders at the University of Idaho beacause the origional design called for a small piston to be made that would compress a fluid, and a pressure could be read. This pressure could then be converted into a torque by calibrating the sensor. However, it was decided that an electronic transducer or load cell of some sort should be installed instead. It fell to the HyRollers to specify the load cell, design a method to mount it, and manufacture the pieces required. In addition, the Dynamometer had never been tested, and when pressurized for the first time after installing packing, it was discovered that the dyno leaked. Liquid Gasket was used to seal most of the leaks, new gaskets were made where the origionals had ripped during the first installation. Next, all the bearings were checked to ensure they would be able to handle the speeds the team would be placing on them (8000 – 10 000 RPM) and deficient bearings were replaced.Finally, it was discovered that the carrier that allows the whole dynamometer to rotate and transmit its tourque to the force transducer, was filled with a very heavy grease. This grease provided an opposing torque to motion that was significant at very low loads, and would have increased the amount of hysterisis in the device by increasing the static friction of the carrier. This grease was removed, and the bearings were placed in the carrier one at a time. This was much more difficult with no grease to hold the carrier bearings, and will cause the ball bearings to spill everywhere if the next person to dis-assemble it is not careful. TAKE NOTE!!Step 3: Modifying the Planetary Gear System left by Jason SagenFigure 11: Jason Sagen’s Proposed Planetary Configuration421957524130This planetary system was left to the University by Jason Sagen’s. It was the topic of his master’s thesis. This was a well designed planetary that allowed access to all three of the possible input/output shafts. He had proposed a certain configuration of the speeder, torquers and IC engine, and the team went in a slightly different direction. Jason originally proposed that the engine be connected to the carrier, the speeder to the sun, and the torque to the output which is on the ring gear. (See Figure 11) This makes sense from a packaging perspective, and the design the HyRollers have developed does not preclude using this in the final hybrid FSAE car, but the ratios make much more sense the way that was laid out in section 6.1.1. In addition to the change in concept, the actual planetary system was modified to accommodate the torques and speeds of actual testing by changing out some bearings for higher quality bearings rated at the required speeds, modifying the internal bushings to stop them from binding, and adding an additional needle bearing at the end of the sun-gear shaft in order to support it more fully. These changes were all fairly minor, but had to be done in order to reduce the friction in the gearbox, and ensure that it could withstand several groups using it for testing.Step 4: Designing and Building the Single Motor Test PlatformOnce the miniature dynamometer and the miniature planetary gearbox were completed it was time to design and construct the single motor test platform. The single motor test platform was designed to test the motor characteristics of the Hacker brushless DC motor using the miniature dynamometer. It required the design and construction of a bracket to hold the motor in alignment with the input shaft on the dynamometer, as well as a bracket to mount the Hall Effect sensor to the motor bracket so the speed of the motor could be determined. A flexible coupler had to be purchased to couple the motor the dynamometer, which allows for a little bit of error in the alignment of the two shafts. Step 5: Testing on the Single Motor Test PlatformThe basic layout of the single motor test is shown below in Figure 12. Because the small controllers supplied by Hacker were not made for use at high-torque low RPM applications, the team ruined three controllers while trying to collect this data. One of the controllers was repaired, but because of the fear of blowing more motors, the team only ran the motor at 46% of the full setting (800ms pulse-width signal sent via the Board of Education). This was being done when the last motor controller blew, but the team was able to gather significantly more data than they were when trying to operate at the full setting. The team has two recommendations to allow future teams to get the full data required. The first is to develop a proper motor controller made to handle these high torque-low speed conditions. And the second is to test the motor while driving it through the planetary gear drive in order to give a reduction. This reduction will allow higher input speeds from the motor, while giving high torque to the water-brake dynamometer.Figure SEQ Figure \* ARABIC 12: Single Motor Testing SchematicOnce all hardware is assembled the load cell must be calibrated before any data can be collected. When calibrating a load cell there is a number of steps that must be taken. Calibrating takes place after the personal computer is set up to read the load cell; in this case MatLab/Simulink is used as explained above. A load cell must be calibrated in order to ensure the starting point begins at zero when initial tests are taken. To begin the calibration weights are added to the plate on top of the load cell and the voltage is read off the computer from the scope in Simulink as shown in Figure 13 below. The yellow flat line seen on the graph is data being collected for the Hall Effect sensor, and the purple is data from the load cell. The yellow Hall Effect line will remain constant since the motor is not spinning during calibration. Figure SEQ Figure \* ARABIC 13: Calibration and Data-CollectionThe two values, mass and voltage, are then recorded. The range of data points we used was from 0-2000 grams and then back down from 2000-0 grams in increments of 100. By taking data points on the from low to high weight as well as from high to low weight allowed us to determine if any significant hysteresis error was present. Interpretation of the data did show that there was a considerable amount of hysteresis. We determined that the main cause of this was from the saddle bearings found directly in front and behind of the dynamometer. These journal bearings support the dynamometers axial rotation generated by a load is exerted on it by the motor, thus putting pressure on the load cell. After plotting the points, voltage vs. mass, we observed a linear regression with a trend line. This data can be seen below in Figure 1. The constants from this equation were taken and entered into the MatLab function window (See Appendix J). The equation of the line is to be used to convert voltage output of the load cell to force for later calculation of the torque. The results of the calibration data collection are shown in Figure 14 below.Figure SEQ Figure \* ARABIC 14: Data Points Collected when doing the Linear CalibrationOnce the load cell is calibrated, the single motor test is ready for data collection and a torque curve can be made. The following is a procedure for collecting data using the single motor test bed. With the valve on the dynamometer open and the water turned on, the MatLab program running, and the motor begins spinning at is specified speed, it is up to the operator to slowly close the water valve on the dynamometer casing the motor to bog down. This causes a torque on the dyno, which is applied through a moment arm to the load cell. What makes this difficult is that the operator must keep the flow rate constant through the dyno to get a smooth set of data. We found this nearly impossible but did get some relevant data after 16 different trials. After considering why it was so difficult to keep continues torque on the motor we came to a conclusion that we felt would provide us much more accurate results. A water break dyno needs a constant amount of water circulating through it to produce an unvarying amount of force on the strain gage. Therefore, instead of trying to control the flow rate to and from the dyno we decided to fill the dynamometer with a specific amount of water. With this change in operation, the scope on the PC should give a much cleaner graph. Essentially this method cuts out any room human error. The down side to this proposal is that the water in the dyno could become very hot if it were to run for a long time. This is why it was originally designed for water to flow in and out keeping the apparatus cool. Lucky for us we are only testing for about ten seconds at a time so heat was not an issue. We were able to get two sets of data using this method before the system crashed due to a blown speed control.Below one can see the difference in trying to control the flow rate as opposed to keeping water level constant. Notice the top two graphs are now in pounds vs. time rather than voltage vs. time by using the calibration equation " y=0.0015x+0.1055." It can also be observed how the corresponding 2 x Revs per minute corresponds to the unrefined strain gage data. Figure 15 below show two examples of raw data collected during the data collection portion of the test.ControllingConstantFigure SEQ Figure \* ARABIC 15: Data Reduction, Isolating Readable DataOnce all 18 data sets where complete our next step was to take the different groups of data and put them into excel. There were three different variables acquired from this experiment, pounds, revolutions, and time. In order to evaluate the data we need to graph pounds and revolutions with respect to a point in time. As seen from above data is only accurate at points that gave us a stable force over time. These specific portions of data were collected and set aside.Converting Revolutions to RPM and in-lbs to torque was nearly the final step into getting our torque vs. RPM curve. The following equation was used to convert Revolutions to RPM;RPM=Rev2-Rev12signalsRev*Time 2-Time 1*60secondsmin To convert in-lbs to Torque merely taking the Average of all the torques in one data set and multiplying by the moment arm, 4.2475 in, where;Torque= Force*DistanceThis analysis resulted in the torque curve below, which matches up very well with predicted values. However, it is important to note that a significant amount of data reduction is needed to get suitable results. Figure 16 shows the final torque vs. RPM curve for the A40-14S Hacker motor tested.Figure 16: Torque Curve for A40-14S Hacker Brushless DC MotorStep 6: Design and Build of the full Dual-Input Test PlatformThe design and build of the dual input test platform was considerably more complex than the single motor test platform. The design of a plate for the two motors and planetary gearbox to be mounted on had to be completed, taking into consideration the proper height of the platform to ensure the closest possible alignment of the planetary output shaft and the dynamometers input shaft. Brackets for holding the motors in place had to be designed as well, taking into account the loading on the shaft of the motor due to the belts and pulleys that would be used to provide the input to the planetary gearbox. These brackets had to be designed with slots in them for the motor shafts to slide into, as well as bearings to provide support for the output shaft. These brackets, like the single motor bracket, had to be designed to support the hall effect sensor so that motor speed data could be taken. These brackets were also made with slots where they mounted to the test bed platform so that the tension in the belts can be adjusted, as well as to accommodate different size pulleys so that multiple gear ratios could be tested. The majority of the work done on the Build of the full dual input test platform can be seen in Appendix E, which contains all of the mechanical drawings of the system.SummaryAll of the work done on this platform was meant to be fully applicable to the full-scale H-FSAE car. The work included adapting previous work done at the University of Idaho to meet these specific requirements, and integrating several existing systems together. In addition to the modifications and upgrades, the team developed a new portion of the test-platform and did a significant amount of engineering problem-solving to make the whole picture come together. There is still more work that needs to be done, but this provides the foundation that will be required by future teams to build on and develop further the concept this team has put forward as an award-winning design. This test-platform can be used to show teams how this systems works and get them up to speed on the project faster by providing a hands-on learning experience. The platform can also be used as a demo model to showcase the achievements of the University of Idaho’s engineering program. The next section provides a list of future work that the HyRollers would recommend pursuing with future teams.Future Work Test-bed specific work (upgrades and wrap-up)Additional work to be done to two motor test-bed (5 weeks)Build and install a water delivery system to the dynamometer which allows operator to precisely control the water level which will allow for more constant loading. Get parts anodized and create clear plastic guards around moving parts.(Optional) Add a flywheel to the dynamometer shaft to give the system “inertia” to overcome, and to recoup braking energy from.(Optional) Add another electric motor to the dynamometer shaft to simulate the additional torque motors that this team propose be placed on the front tires to better gather braking energy.Additional work to be done to two motor test-bed data acquisition system (2 to 3 weeks)Write new Basic Stamp program for future torque testing.Improve Simulink program to automatically calculate torque and RPM real time without causing lag or aliasing.Determine how to initiate the Basic Stamp program to run the motors using Simulink. This will allow testing to be initiated with a single input.Future Testing that should be performedAdditional motor torque testing of the A40-14S and A40-10L (2 to 3 weeks)Additional testing of the A40-14S motor is needed at several different speeds. The data collected by HyRollers was done at a pulsout setting of 800 (46.3%). The motor operates at pulsout settings between 497 and 1152. This should be done through the planetary to allow higher torque on the Dynamometer and higher operation speeds of the motor being tested.Similar testing for the slightly larger A40-10L also needs to be conducted to determine a method of scaling torque curves of these smaller motors to the larger A200-20S. for use in the actual racecar.Extrapolation of torque curve for A200-20S using data collected from A40 testing.Determination of operating characteristics of multiple inputs to the planetary drive system (6 to 8 weeks)Develop a test regimen for dual input system.Conduct tests on dual input system.Add the Howell V4 engine to the platform (Appendix C).Develop control for the V4 (starting sequence, maintain power).Further Development of Hybrid system using the Scale Test-StandDesign of power management hardware and software to be vetted on the two motor test-bed (1 to 2 Semesters)Electrical Engineering team to develop the complex control system required.Designation of full-scale hybrid drivetrain components (1 Semester)Using results from scale model component testing, recommend full scale components.Provide updated budget for full scale components.Determination of final planetary drive ratios (1 to 2 weeks)Use a variety of pulleys and belts to vary drive ratios.Using results from various drive ratios determine best ratio suitable for race application.Final Design of hybrid drivetrain (1 to 2 Semesters)Specify layout of drivetrain in vehicle and manufacture.Make final recommendation for hybrid vehicleWork CitedAn, F., F. Stodolsky, and D. Santini. 1999. “Hybrid Options for Light-Duty Vehicles.” SAE Technical Paper 1999-01-2929.2.DaimlerChrysler Media Services. 2007 Dodge Durango: Bold New Look With Class-leading Power, Performance and Convenience. Dodge Press Release. [Online] April 5, 2006. [Cited: July 13, 2008.] Motors. General Motors Alternative Fuel Vehicle Overview. General Motors. [Online] July 13, 2008. [Cited: February 4, 2008.] , K. 1998 “Evaluation of Toyota Prius Hybrid System (THS).” Ann Arbor, MI?: U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Mobile Sources, Advanced Technology Support Division, Technology Development and Support Group.. Hybrid Tachnology Cars. . [Online] July 13, 2008. [Cited: July 13, 2008.] . Kimura, A., I. Ando, and K. Itagaki. 2005. “Development of Hybrid System for SUV.” SAE Technical Paper 2005-01-0273.Levin, M., S. Kozaredar, J. Chottiner, E. Maucher, A. Karamavruc, and R. Shankland. 2002. “Hybrid Powertrain with an Engine-Disconnecting Clutch.” SAE Technical Paper 2002-01-0903.Lexus, a Division of Toyota Motor Sales, U.S.A., Inc. Lexus All Models. Lexus: New Luxury Cars and SUVs from Lexus USA. [Online] July 13, 2008. [Cited: July 13, 2008.] North American Operations. 2008 Mazda Tribute Hybrid. Mazda USA. [Online] July 13, 2008. [Cited: July 13, 2008.] , K., M. Yamazaki, and J. Tokieda. 2004. “Development of New-Generation Hybrid System THS II Drastic Improvement of Power Performance and Fuel Economy.” SAE Technical Paper 2004-01-0064.Nedungadi, A., S. McBroom, C. Roberts, J. Harris, and M. Walls. 2002. “A Parallel Hybrid Powertrain for a Motorcycle Application.” SAE Technical Paper 2002-32-1779. Nissan Motor Co., Ltd. 2008 Nissan Altima Sedan Models. Nissan USA. [Online] July 13, 2008. [Cited: July 13, 2008.] , J. S. 2008. “Hybrid-Electric Planetary Drivetrain Design for Automotive Applications.” Moscow: University of Idaho.14.Santini, D., A. Vyas, and J. Anderson. 2002. “Fuel Economy Improvement via Hybridization versus Vehicle Performance Level.” SAE Technical Paper 2002-01-1901.15.Simic, D., H. Guiliani, C. Kral, and F. Pirker. 2006. “Simulation of Conventional and Hybrid Vehicle Including Auxiliaries With Respect to Fuel Consumption and Exhaust Emissions.” SAE Technical Paper 2006-01-0444.16.Thayer School of Engineering at Datmouth. [Online] July 13, 2008. [Cited: July 13, 2008.] http:// formula-.17.Toyota Motor Corporation. Toyota Vehicles. Toyota Home Page. [Online] February 4, 2008. [Cited: July 13, 2008 .] , J., H. Husted, and K. Rajashekara. 2001. “Comparative Study of Hybrid Powertrain Strategies.” SAE Technical Paper 2001-01-2501.19.Toyota to race Hybrid in Le Mans 2010, Inventor Spot [Cited: Dec 12, 2008 .] of AppendicesARoad Loading Model Developed in SimulinkBPSAT DocumentationCHowell V-4 DocumentationDComponent DatasheetEDrawing Package of Test PlatformFCalculations to Find the Expected Force on the Force TransducerGWiring Diagram for Single Motor TestingHWiring Diagram for Double Input TestingIBasic Stamp DocumentationJSimulink Data Acquisition Program DocumentationKExperimental ProcedureLDatabase OrganizationMTotal ExpensesAppendix A Road Loading Model Developed in Simulink Proposed Hybrid Performance BenchmarkScreenshot of the Simulink program that would be used to predict the performance of the vehicle utilizing only the YZ250F engine, to serve as a performance benchmark.Subsystem: Speed iterationSimulink program that could be used to calculate the desired speed for an imaginary racecourse. Subsystem : Road Layout modelSimulink program used to simulate the driving cycle for an imaginary track.Subsystem: Hybrid being modeledSimulink model made in PSAT of a potential Hybrid drivetrain to be used.Appendix BPSAT DocumentationAppendix CHowell V-4 DocumentationHowell V-4 Engine DescriptionThe engine is 1.95 Cu. In. (32cc) in displacement. Cylinders are 90 degrees apart in order to have the engine balanced for vibration free running. The cylinder bore is .875" and the piston stroke is .812". The cylinder banks are not staggered and robust knife and fork connecting rods are used. The multi segment built-up crankshaft and twin cam shafts are amazingly easy to make. A Hall Effect distributor is driven off the end of one of the cam shafts. The distributor body is linked to the throttle arm for spark advance/retard with the throttle setting. The throttle is my newest proven 2-jet design with an oiled foam air cleaner. Pressure lubrication to the rod ends is by an external gear oil pump which feeds oil through the drilled crankshaft. There is an oil pressure adjuster and an oil pressure gauge port. The engine is water cooled using my unique magnetic drive water pump which has no seals to leak, and a proper looking and effective shop made radiator. The fan blade shroud insures that the 5 curved blade fan actually pulls air through the radiator fins and not just circulate the air around behind the radiator as would otherwise result without one.There are ball bearings on the crankshaft, timing gears, camshaft, distributor, rocker arms, water pump, oil pump and an shaft. All external parts are sealed using "O" rings which prevent any oil seepage from the engine. A crankcase vent/check valve maintains negative crankcase pressure. A dipstick is provided to monitor the oil level and also an easy to get to oil drain plug.Checking the RPM with a Laser Digital Tachometer on initial test runs show an idle speed of approximately 1,000 RPM and a top speed of 6,500+ RPM. Both of these are expected to improve after the engine is fully broken-in.Every feature of the engine as shown is included in the plans set including 2-Jet Throttle, Air Cleaner, Water Pump, Oil Pump, Hall Effect Distributor, Radiator, Skid, Etc. Optional high quility "lost wax investment" intake and exhaust manifold castings are available for builders wanting them. The plans set consists of 65 detailed sheets of quality laser printed CAD drawings and 5 sheets of Construction Notes.Specifications:Flywheel Diameter: 3"Height on Skid: 7.07"Overall Width: 5.75"Length w/Radiator: 7.6"Information Taken from: Application for Hybrid FSAEThe final design for the drivetrain will include the use of an internal combustion engine. The Howell V-4 that was manufactured in the machine shop will be a great model for our scale test stand. While the V-4’s characteristics will be different from the single cylinder Yamaha engine to be used in the full scale, it will do a better job of representing the transient response of an internal combustion engine than any electric motor could. 12509517145320865517145Actual HowellRendered HowellRendered Test Bed with HowellAppendix DComponent Datasheets(For More Detailed Information, see the Electronic Copies Saved By Team)A40-10L??Developed for 5-6 lb Sport, Scale, and 3D-Aerobatic models with 3-6 cell LiPoly batteries. The 14-pole outrunner type design creates amazing torque, therefore larger direct drive props can be used without the need for a gearbox. These motors feture oversized bearings, curved neo magnets and high efficiency stator design. Additionally you will find a special large concentric bearing located in the rotor to support the rotating mass during extreme 3D. A cooling fan is mounted on the motor for closed cowling or helicopter applications to provide proper motor cooling. Idle Current:2.7 ampsOperating Current:45 ampsPeak Current: 55 ampsPeak Watts:1100Resistance:0.018RPM / V:500Battery:3-6 LiPoShaft Diameter:5.0mmShaft Diameter w/GBn/aWeight350g (12.3oz)Weight w/GBn/aA40-10S??Developed for 4-5 lb Sport, Scale, and 3D-Aerobatic models with 3-5 cell LiPoly batteries. The 14-pole outrunner type design creates amazing torque, therefore larger direct drive props can be used without the need for a gearbox. These motors feture oversized bearings, curved neo magnets and high efficiency stator design. Additionally you will find a special large concentric bearing located in the rotor to support the rotating mass during extreme 3D. A cooling fan is mounted on the motor for closed cowling or helicopter applications to provide proper motor cooling. Idle Current:3.6 ampsOperating Current:40 ampsPeak Current: 50Peak Watts:900Resistance:0.012RPM / V:750Battery:3 - 5 LiPoShaft Diameter:5.0mmShaft Diameter w/GBn/aWeight265g (9.3oz)Weight w/GBn/aSpecificationsOperating Current (A):55Servos:2-6Peak Current (A):n/aDimensions:2.9"x1.1"x.4"Battery:2-6 LiPo / 8-18 NiMhWeight:1.6 ozBEC?:YesX-55 SB ProX55SBPROX-70 SB ProX70SBPRO?? SpecificationsOperating Current (A):70Servos:2-6Peak Current (A):n/aDimensions:2.9"x1.1"x.5"Battery:3-6 LiPoWeight:1.9 ozBEC?:YesConnection technologyEVC007ADOAH043MSS0002H04SocketFor sensors withM12 connectorFree from siliconeFree from halogengold-plated contacts?Electrical design DC PNPOperating voltage [V]10...36 DCCurrent rating [A]4?Design angledOperating temperature [°C]-25...90Protection IP 67 / IP 68 / IP 69K, IIMaterial body housing: TPU (urethane) black transparent; sealing: VitonMaterial nut brass; nickel-platedTightening torque for knurled nut [Nm] 0.6...1.5Function display Power LEDgreenSwitching status LED2 x yellowConnection PUR cable / 2 m; 4 x 0.34 mm? (42 x ? 0.1 mm); ? 4.9 mm; halogen-freeSheath color blackDrag chain suitability Bending radius for flexible applications:?min. 10 x cable diameterTravel speed:?max. 3.3 m/s for a horizontal travel length of 5 m and max. acceleration of 5 m/s?Bending cycles:?> 5 million?Wiring Core colors BKblackBNbrownBUblueWHwhiteifm efector, inc. 782 Springdale Drive, Exton, PA 19341 ?—? We reserve the right to make technical alterations without prior notice. ?—? US ?—? EVC007 ?—? 27.07.2006-895350-914400Appendix E Drawing Package of Test Platform Appendix F Calculations to Find the Expected Force on the Force TransducerMotor Calculations and Force transducer sizingfor HyRollersPhase 1 and Phase 2 hybrid driveline mockupJedidiah BartlettOctober 10 2008Problem Statement:Determine from a limited amount of given information the forces that exist throughout the bench-top hybrid driveline for both a single motor test, and the completed configuration that includes the planetary power differential. This analysis should be summarized with a range of forces that the force transducer on the dynamometer will see in the two configurations.The following equations show how the HyRollers estimated the amount of torque that will be applied to the Dynamometer. There were no exact figures given by Hacker Brushless about starting torque or maximum torque of the motor. From their website, they did have the performance numbers of the motor when loaded with two separate props. These cases are shown below directly from the website.MotorPropLiPoVoltAmpRPMPower (Watt)ESCA40-14SAPC-E 14x104S14.042.16496590X-55-SB-ProA40-14SAPC-E 13x6.55S17.538.28598669X-55-SB-ProThe physical data for the motor is found below:SpecificationsRPM/VWeightShaft DiameterBatteryOperating Current(Amps)Peak AmpsPropPeak WattsA40-14S530265g (9.3oz)5.0mm3-5 LiPo40 amps70APC-E 13x6.5 to 14x10900And finally, the motor constants are shown below.MotorRPM/V (Kv)Idle Current (lo)Resistance (Ri)A40-14S5302.4 amps0.021This motor was selected because it is capable of high power, (being a 14 pole machine). But it also had the lowest RPM/Volt rating; which means that it spins the slowest of the A40 series motors for the same power output. Because we are more concerned about torque, and are already having trouble accommodating the high speeds that these motors are capable of, we felt it prudent to select the motor designed for torque rather than speed.Using the case supplied above that develops more power, the torque at that point can be calculated using the formula: Where T is torque in in-lbf. Rearranging this equation and to solve for T, and substituting the value for HP in converted from Watts:TIf this number seems low, it is because it is. The torque curve for a brushless DC motor typically exhibits the following general shape:669 W6.576 in-lbBecause this prop allows a very high RPM, the operating point will be near the low end of the torque curve. The HP is the area contained within the dashed box. 669 WRPM8598 RPMFor the second point given, the prop is larger, and as such requires a higher torque. This means that the equilibrium point will move to the left on the graph, allowing the motor to supply more torque, but at a slower speed.590 W7.676 in-lbTBecause this prop allows a lower, the operating point will be closer to the high end of the torque curve. The HP is the area contained within the dashed box. RPM6496 RPMBecause there is no way to know the point where the torque plateaus, we will assume it to be 1/5 of the maximum RPM (Based on the curves from UQM motors , which provide large brushless DC motors for automotive applications). The maximum RPM is assumed to be about 10 000 RPM when there is no load other than bearing friction, but the theoretical speed will most likely be double that. These are both simply engineering estimates, but they should put us in the right ballpark for sizing our torque sensor.Therefore, the maximum torque will occur at about 4000 RPM. Using the two points given, a fit line can be computed, and the maximum possible torque calculated.The fit line is calculated in this manner:Using the linear form Y = mx+b (where m = Slope), and knowing one of the points, we can solve for b:Substituting b and m into the linear standard equation form we find that the equation for the fit line is:Solving this equation at Torque = 0 will yield the theoretical top speed. However, the actual top speed is substantially lower. Consider the graph from UQM’s website for their 200hp (PowerPhase 150) motor as an example. The theoretical RPM works out to be 21201 RPM, which is about twice what we estimated the actual maximum RPM to be.We can calculate about where the discontinuity in the graph is on our setup by assuming that the maximum torque will be supplied at 1/5th of the maximum operating speed, which we have assumed to be 10 000 RPM; making the maximum torque occur at 2000 RPMThis of course assumes that the expression after the discontinuity is linear. That does seem to be the case on larger motors of this configuration, but because the bearing-drag on smaller motors impacts the hp available more-so on smaller motors than larger, this assumption may be faulty. The team has found smaller DC brushless permanent magnet motors with both linear, and quadratic sloped portions.Because of that, we will use the y-intercept of the sloped line (11.06) and round it to one lb-ft of torque (12 in-lb) to size the load sensor.For Phase 1 of the design process where one motor is being used, the load on the cell is simple to calculate:It will have an upper limit of And a predicted value of: However, the derivation for the force when both motors are hooked up is more complicated, and is laid out on the following pages.The Planetary system can be represented symbolically as shown below, with the effective radiuses:In order to calculate the force on the output carrier, which attaches all of the planetary gears together, one must perform a force analysis on one of the planetary gears. Remember that this system would have exactly the same ratio if there was only one planetary gear. So the analysis below assumes that all of the forces act on a single gear. This can also be explained by saying that, the forces will be evenly distributed around the four gears, and the sum of the 4 gears forces act on the ring and the sun, so ? of the forces, times 4 gears, is the same as just having one gear.A FBD of this “single virtual planetary gear” is shown below:FringFsunFcarrier From this it is simple to see that the Force exerted on the pin of the planetary gear will equal the applied forces from the sun and the ring. That means that the torque load applied through the carrier shaft will equal the driving torques from the motors through the ratio of their moment arms (effective diameters)The forces are calculated here. = 32.4 in- lbfNow that the torque is known, the force on the Dynamometer arm can easily be calculated because it is known that the moment arm on the dynamometer is 4.25inForce on Dyno sensor: (For the two-motor configuration in Phase 2 of the project.)Using the predicted value of the torque of the motor rather than the maximum using the above procedure yields:The results of the above calculations are tabulated below.Force expected to be exerted on the Dyno sensor(in lbf)Upper limit based on y-intercept of the fit lineExpected value based on 2000 RPM value of fit lineFor Phase 1 when testing only 1 motor direct to dyno2.823 lbf2.35 lbfFor Phase 2 when testing two motors coupled together through planetary gears7.62 lbf6.35 lbfBecause of the above data, we will be purchasing a 10lbf sensor will ensure that the scale of the instruments is appropriate for the size of motors being used. This can be ordered from Omega () with the following part number. It may also be advisable to purchase a 20 lb load cell to be able to better measure transients without maxing it of FormLCL-010 ?? $69.00??? Thin beam load cell with 10 pound capacity ? qty. ?LCL-020 ?? $69.00??? Thin beam load cell with 20 pound capacity ? qty. ?LCL-040 ?? $69.00??? Thin beam load cell with 40 pound capacity ? qty. ?LCL-CL1 ?? $15.00??? Mounting kit for LCL-113G thru LCL-816G (required) ? qty. ?LCM-CL1 ?? $30.00??? Mounting kit for LCL-005 thru LCL-040 (required) ? qty. Bottom of FormIt should also be noted here that we are assuming the Dynamometer will be able to handle these speeds. This is something else we are unsure of.Appendix GWiring Diagram for Single Motor TestingAppendix HWiring Diagram for Double Motor Input TestingAppendix IBasic Stamp DocumentationBasic Stamp Single Motor servo controller This program was the first program used and illustrates the basic functioning of the program and how to control a servo motor. It bears mentioning that these output pulses generally are sent to give a servo motor a specific position over the range of motion. This range is usually from 496ms to signal a move to 0 degrees, and 1150 to signal a move to 90 degrees, and any percentage between to signify positions in between. In this case however, the percentage between these two pulse-widths sets the desired power output from 496 being 0 output, to 1150 being full output. The controllers provided by Hacker require that when they are first powered, that they receive a pulse-width output of between 400 and 496 in order to initialize the controller. This prevents the person plugging it in from having the motor begin to turn because the controller was left on. ‘ Basic Stamp Single Motor servo controller' {$STAMP BS2}' {$PORT COM6}svo CON 15 'I/O contro for the servomotorpulse_width VAR Word ' pulse_width to control the speed of the servomotorrun VAR Byteiter VAR Wordcyc:run = 1GOTO mainMain:IF run > 1 THEN prepareIF run <= 0 THEN finishGOTO prepareprepare:LOW 11FOR iter = 0 TO 1000PULSOUT svo,0NEXTPAUSE 1000FOR iter = 0 TO 1000PULSOUT svo,400NEXTGOTO outpulseoutpulse:pulse_width = 1200 ' actual setting of the motor because the serial option required ' Too much troubleshooting.IF ((pulse_width < 500) OR (pulse_width > 1501)) THEN Main'Check that pulse sent to servomotor is 'within the desirable rangeHIGH 11FOR iter = 0 TO 255PULSOUT svo,pulse_widthNEXTPAUSE 10run = run - 1GOTO Mainfinish:ENDBasic Stamp Single Motor Servo Controller with Ramp-up It was found that the motors struggled under high-load when given the full throttle setting that was desired, and the team discussed the possibility of ramping the motor up to the desired throttle setting. The below program was used to do this. It also incorporates the serial in and out commands used to receive data from Matlab. Further instructions can be found on how to interface Matlab and Basic Stamp in the paper titled “Matlab-Based Graphical User Interface Development for Basic Stamp 2 Microcontroller Projects” by Yan-Fang, Saul Harari, Hong Wong, and Vikram Kapila (Dated September 2003). The HyRollers team could not get the Matlab portion to work because Matlab would not always recognize the serial ports. It would be very nice to initiate the starting of the motors from Matlab, so these portions were left in as comments so that future teams can build on this work.' {$STAMP BS2}' {$PORT COM6}svo CON 15 'I/O control for the servomotor'serial CON 16 'PIN 16 is used for serial communication, since the BS2 'has a line receiver on its SIN pin (Rpin 16). All of BS2's 'I/O pins can receive RS-232 data serially. To utilize the build-in 'serial port ser Rpin TO 16. TO use other I/O pins FOR serial 'communication, a 22Kohm resistor is needed.'baud CON 84 'The BS2 and the PC must be configured on the same baud rate to 'communicate successfully on the serial port. In this experiment 'the baud rate is 9600. BS2 automatically converts this constant of '84 to a baud rate of 9600 when it executes the serin commandpulse_width VAR Word ' Pulsewidth to control the final speed of the servomotor.pulse VAR Word 'Used to increment the speed during ramp-uprun VAR Byte 'Designates the run modeiter VAR Wordaccel VAR Wordlight VAR Byte ' toggles light position during accelerationhold_cyc VAR Word ' This variable controls the amount of time that the motor holds speedspeed VAR Wordcyc:'-------------Set Points that can be changed------------------------ 'This section of code required too much de-bugging to get to work. There were I/O problems 'That were not easy to resolve between Matlab and the Basic Stamp. I believe this may have 'been because of the NI 6008 that was used to collect data. However, if the serial communication 'could be used, it would allow Matlab to control the operation/initialization as well as 'data acquisition 'SERIN serial,baud,[DEC pulse_width] 'Receives desired speed as the duration 'as the duration of a pulsout commandspeed = 50 'Set between 496 and 1150'This was originally designed so that the "speed" would be a setting between 0 and 100%'The operator would then set this variable between 0 and 100 to give the throttle position in %'see the section under "set_pulse" for more informationhold_cyc = 1000 'this determines the number of cycles that the speed is held foraccel = 50 'The lower the acceleration number, 'the faster the motor will ramp-up'set so that it can equal 450+n*accel, otherwise hold_low will be used to ramp-down'in other words, use a rounded number to hold motor at max setting'and use a non-rounded number to ramp-down the motor without holding it at max'------------Program------------------------------------------------IF speed > 0 THEN set_pulsepulse_width = 450IF speed <= 0 THEN Mainset_pulse:'This didn't seem to work properly, when this line was added, the program would no longer'operate at all. I believe that this microchip cannot handle all the operations required'by this conversion from a % of speed to a pulse width.'pulse_width = ( ( speed / 100 ) * 650 ) + 500 ' actual final setting of the motor,pulse_width = speedrun = 1 ' Begins programGOTO MainMain: IF run = 1 THEN prepare IF run = 2 THEN ramp_on IF run >= 3 THEN finish ENDprepare: HIGH 11 FOR iter = 0 TO 500 PULSOUT svo,0 NEXT PAUSE 1000 LOW 11 light = 0 pulse = 450 FOR iter = 0 TO 1000 PULSOUT svo,pulse NEXT run = 2 GOTO Mainramp_on: IF (light = 1) THEN ramp_off IF pulse = pulse_width THEN hold IF ((pulse_width < 400) OR (pulse_width > 1501)) THEN Main'Check that pulse sent to servomotor is 'within the desirable range HIGH 11 light = 1 pulse = pulse + 50 FOR iter = 0 TO accel PULSOUT svo,pulse NEXT GOTO ramp_offramp_off: IF pulse = pulse_width THEN hold IF ((pulse_width < 400) OR (pulse_width > 1501)) THEN Main'Check that pulse sent to servomotor is 'within the desirable range LOW 11 light = 0 pulse = pulse + 50 FOR iter = 0 TO accel PULSOUT svo,pulse NEXT GOTO ramp_onhold: FOR iter = 0 TO hold_cyc PULSOUT svo,pulse_width NEXT run = 3 GOTO Mainfinish: LOW 11 light = 0 PAUSE 10 ENDBasic Stamp Single Motor servo controller This final program was used to step both of the motors through different throttle settings when both motors were hooked up through the planetary system. This allows the user to set three different “stages” where the motors can be set to different settings and the effects observed. All of the settings are set in Main in this program.Because of the limited number of variables available, this program utilizes all of the available variables. Making a longer program with more stages would require that the values be entered within each loop and not be assigned to separate variables.' {$STAMP BS2}' {$PORT COM6}LED CON 11 ' Port setup for LED (connected in series with a resistor to limit current)svo1 CON 14 'I/O control for the servomotorsvo2 CON 15pulse1 VAR Word 'Used to set the pulse width of servo 14pulse2 VAR Word 'Used to set the pulse width of servo 14run VAR Byte 'Designates the run modeiter VAR Wordlight VAR Byte ' Toggles light position during accelerationspeed11 VAR Word ' This variable controls the speed of the motor (set 1 - 650) stage 1speed21 VAR Word ' This variable controls the speed of the motor (set 1 - 650) stage 1time1 VAR Word ' This is used to determine the number of times to send stage 1 signalspeed12 VAR Word ' same as above, but for stage 2speed22 VAR Word ' same as above, but for stage 2time2 VAR Word`' same as above, but for stage 2speed13 VAR Word ' Stage 3speed23 VAR Word ' Stage 3time3 VAR Word ' Stage 3Main: 'Stage 1 speed11 = 50 ' set the speed of the motor on servo 14 between 1 and 650 for stage 1 speed21 = 50 ' set the speed of the motor on servo 15 between 1 and 650 for stage 1 time1 = 5000 ' Set the number of times to send the pulse-out signal. 'The time setting indirectly controls the duration of the stage. 'Stage 2 speed12 = 0 speed22 = 100 time2 = 5000 'Stage 3 speed13 = 100 speed23 = 100 time3 = 5000 'Flash the LED to let operator know that the program is about to start. FOR iter = 0 TO 10 TOGGLE LED PAUSE 200 ' Pause for one half a second NEXT GOTO prepareprepare: ' The controllers have to receive a 0 pulse when initially plugged in, and then a 400ms ' pulse in order to initialize. The motor will beep while the controller is initializing. TOGGLE LED FOR iter = 0 TO 1000 PULSOUT svo1,0 PULSOUT svo2,0 NEXT PAUSE 1000 FOR iter = 0 TO 1000 PULSOUT svo1,450 PULSOUT svo2,450 NEXT run = 1 GOTO run1outpulse:' This is the section of the program that steps through all of the 3 different run states.' additional states could be added, but there are too many variables to add them without' reconfiguring the operation of the program. the BS2 chip cannot track any more variables. TOGGLE LED FOR iter = 0 TO 500 PULSOUT svo1, pulse1 PULSOUT svo2, pulse2 NEXT run = run + 1 IF run = 2 THEN run2 IF run = 3 THEN run3 IF run = 4 THEN finishrun1:'Stage 1 output pulse1 = speed11 + 500 pulse2 = speed21 + 500 GOTO outpulserun2:'Stage 2 output pulse1 = speed12 + 500 pulse2 = speed22 + 500 GOTO outpulserun3:'Stage 3 output pulse1 = speed13 + 500 pulse2 = speed23 + 500 GOTO outpulsefinish:' Turn the LED off from whatever state it is in, and end program.' Once downloaded, the program can be run again by pushing the reset button on the' Board of education LOW LED ENDAppendix JSimulink Data Acquisition Program DocumentationThe Failed Simulink ModelWorth re-visiting in the future:-404495178435This was the first Simulink model that was developed in order to measure the RPM of the test-stand in real-time.The model counts the number of pulses ( 2 per revolution), and then every half a second divides the elapsed number of pulses from the previous count by the elapsed time. This model was very slow, and could not keep count of the pulses properly because, even though the sample time of the Digital Input was adequate, the computation time for the above model slowed the system down. The multiplication blocks act as logic AND gates and multiply the number being passed through by either a 1 or a zero in order to maintain the proper output. The tall, narrow addition blocks act as OR gates. Because only one of the two inputs is active at a time (not being multiplied by zero), it passes only the active value through the gate. This model could be improved by not keeping track of the elapsed time, and assuming that the time elapsed will be equal to the time that the count is evaluated at ( 0.5 seconds). It was originally thought that the two would be slightly different, but after observing this model, it became evident that the samples are taken as close to the entered sample time as could be expected, and that tracking the actual elapsed time provided no benefit, but only further slowed the computations. The biggest time-limiter though seemed to be the use of the memory blocks, which are labeled above as “read ‘variable’” and “write ‘variable’” it was when these were introduced to the system that the simulation began to lag to the point of being useless for actual data acquisition. If this model could be made to work, with a faster computer and finding a faster way to store variables, then the output of any test could be plotted as RPM and Torque directly in real time and facilitate more complex simulations and dynamic response testing. The measured strain-gauge could be translated directly into torque by use of a user-defined equation in order to accomplish this.The Simulink Model Actually UsedThe second Simulink Model that was used was much simpler, and was designed to collect only the raw data, and an analysis could be made afterwards to extract the desired information. This requires a substantial amount of extra work on the part of the person taking the data, but it was the only way to get good, clean data in the time required.This model simply counts the number of pulses received from the hall-effect sensor, and stores the value along with the measured force. The only calculation done here is the calibration done to change the voltage received from the force transducer into pounds of force. Appendix KExperimental ProcedureEquipmentFirst and for most to make this experiment possible required a number of pieces of equipment and shown and explained below. HP DC power supply → 0-35 Volts/ 0-60Amps → Model # 6673APowers MotorHP Triple Output DC Power supply → 0-6V, 5A/0-±25V, 1APowered each sensor “load cell and Hall Effect”Toshiba Personal ComputerRuns MatLab and Simulink controlling motor and reading data from Hall Effect and Load Cell.Hall effects Sensor → Cherry GS10201Reads the potential difference across an electrical conductor in which an electric current flows in the presence of a magnetic field. The conductor in our case was a steal flag much like in the shape of a football measuring 2 pulses for every revolution.X-70 Hacker Speed ControlAllows computer to control speed of motor12-Bit National Instruments DACU → USB compatibleConnects controls to computerParallax Board of education with B52 chipOmegadyne inc. Load Cell → 0-10 lbs.Outputs a voltage vs. time to Personal computerThe basic layout of the single motor test is shown below in Figure 12. Because the small controllers supplied by Hacker were not made for use at high-torque low RPM applications, the team ruined three controllers while trying to collect this data. One of the controllers was repaired, but because of the fear of blowing more motors, the team only ran the motor at 46% of the full setting (800ms pulse-width signal sent via the Board of Education). This was being done when the last motor controller blew, but the team was able to gather significantly more data than they were when trying to operate at the full setting. The team has two recommendations to allow future teams to get the full data required. The first is to develop a proper motor controller made to handle these high torque-low speed conditions. And the second is to test the motor while driving it through the planetary gear drive in order to give a reduction. This reduction will allow higher input speeds from the motor, while giving high torque to the water-brake dynamometer.Figure SEQ Figure \* ARABIC 12: Single Motor Testing SchematicOnce all hardware is assembled the load cell must be calibrated before any data can be collected. When calibrating a load cell there is a number of steps that must be taken. Calibrating takes place after the personal computer is set up to read the load cell; in this case MatLab/Simulink is used as explained above. A load cell must be calibrated in order to ensure the starting point begins at zero when initial tests are taken. To begin the calibration weights are added to the plate on top of the load cell and the voltage is read off the computer from the scope in Simulink as shown in Figure 13 below. The yellow flat line seen on the graph is data being collected for the Hall Effect sensor, and the purple is data from the load cell. The yellow Hall Effect line will remain constant since the motor is not spinning during calibration. Figure SEQ Figure \* ARABIC 13: Calibration and Data-CollectionThe two values, mass and voltage, are then recorded. The range of data points we used was from 0-2000 grams and then back down from 2000-0 grams in increments of 100. By taking data points on the from low to high weight as well as from high to low weight allowed us to determine if any significant hysteresis error was present. Interpretation of the data did show that there was a considerable amount of hysteresis. We determined that the main cause of this was from the saddle bearings found directly in front and behind of the dynamometer. These journal bearings support the dynamometers axial rotation generated by a load is exerted on it by the motor, thus putting pressure on the load cell. After plotting the points, voltage vs. mass, we observed a linear regression with a trend line. This data can be seen below in Figure 1. The constants from this equation were taken and entered into the MatLab function window (See Appendix J). The equation of the line is to be used to convert voltage output of the load cell to force for later calculation of the torque. The results of the calibration data collection are shown in Figure 14 below.Figure 14: Data Points Collected when doing the Linear CalibrationOnce the load cell is calibrated, the single motor test is ready for data collection and a torque curve can be made. The following is a procedure for collecting data using the single motor test bed. With the valve on the dynamometer open and the water turned on, the MatLab program running, and the motor begins spinning at is specified speed, it is up to the operator to slowly close the water valve on the dynamometer casing the motor to bog down. This causes a torque on the dyno, which is applied through a moment arm to the load cell. What makes this difficult is that the operator must keep the flow rate constant through the dyno to get a smooth set of data. We found this nearly impossible but did get some relevant data after 16 different trials. After considering why it was so difficult to keep continues torque on the motor we came to a conclusion that we felt would provide us much more accurate results. A water break dyno needs a constant amount of water circulating through it to produce an unvarying amount of force on the strain gage. Therefore, instead of trying to control the flow rate to and from the dyno we decided to fill the dynamometer with a specific amount of water. With this change in operation, the scope on the PC should give a much cleaner graph. Essentially this method cuts out any room human error. The down side to this proposal is that the water in the dyno could become very hot if it were to run for a long time. This is why it was originally designed for water to flow in and out keeping the apparatus cool. Lucky for us we are only testing for about ten seconds at a time so heat was not an issue. We were able to get two sets of data using this method before the system crashed due to a blown speed control.Below one can see the difference in trying to control the flow rate as opposed to keeping water level constant. Notice the top two graphs are now in pounds vs. time rather than voltage vs. time by using the calibration equation " y=0.0015x+0.1055." It can also be observed how the corresponding 2 x Revs per minute corresponds to the unrefined strain gage data. Figure 15 below show two examples of raw data collected during the data collection portion of the test.ControllingConstantFigure 15: Data Reduction, Isolating Readable DataOnce all 18 data sets where complete our next step was to take the different groups of data and put them into excel. There were three different variables acquired from this experiment, pounds, revolutions, and time. In order to evaluate the data we need to graph pounds and revolutions with respect to a point in time. As seen from above data is only accurate at points that gave us a stable force over time. These specific portions of data were collected and set aside.Converting Revolutions to RPM and in-lbs to torque was nearly the final step into getting our torque vs. RPM curve. The following equation was used to convert Revolutions to RPM;RPM=Rev2-Rev12signalsRev*Time 2-Time 1*60secondsmin To convert in-lbs to Torque merely taking the Average of all the torques in one data set and multiplying by the moment arm, 4.2475 in, where;Torque= Force*DistanceThis analysis resulted in the torque curve below, which matches up very well with predicted values. However, it is important to note that a significant amount of data reduction is needed to get suitable results. Figure 16 shows the final torque vs. RPM curve for the A40-14S Hacker motor tested. Figure 16: Torque Curve for A40-14S Hacker Brushless DC MotorAppendix LDatabase OrganizationKnowledge Management for Future TeamsOrganizational Layout of T-drive-5981701195705For organizational purposes of our Research, Minutes, Resources, Project Tracking, and all other useful information we created a navigation tree seen below in Figure 12. This tree will allow for future teams, as well as our current team, to locate specific information with ease. Each specific folder contains relating information that we will now explain in more detail. As seen from Figure 12 there are seven main folders; Research/Documentation, Minutes/Agendas, Project Tracking, Team Toolbox, Website, Final Presentations & Papers, Resources, Pictures, and Single Motor Data. Due to growing research and progress this tree will be updated regularly.T-drive Organization TreeResearch/Documentation is one of the largest of the seven folders. This folder contains a number of different things that must be considered when designing a drive train. In this file you will find a wealth of information on Accumulators, DC & AC motors, Gas Engines, Line Diagrams of different configurations, Other Universities Information, Road Load Model ideas, Track Information, and Purchased. In the Accumulators folder you will find research on different kinds of accumulators; such as, Maxwell Ultra Capacitors, Tavirm Super Capacitors, Nesscap capacitors, and Li-ion Batteries. In this file you will also find a presentation of different accumulator packages.There are a number of DC & AC motors on the market. We found, and researched, what we establish to be the best of these motors. This folder contains three different kinds of motors; Perms, Hackers, and a water cooled motor. Along with these there is also a motor selection guide that may come in handy for the future as well. Gas Engines is the folder that contains a power vs. torque curve that we are using as a benchmark for the overall hybrid vehicle. We know that our final hybrid vehicle needs to perform better than the YZ250F on its own. Also you will find some information on the Saturn hybrid.Line Diagrams were a very large part of our research in trying to find the right configuration. This folder has four subfolders which consist of complex, parallel, series, and Stripped FSAE Frame. Each of these folders is fairly self explanatory. The complex, parallel, and series folders not only contain their line diagrams but also a weight cost analysis (Excel files) of different possible configurations. The stripped FSAE frame is a solid model of the 07-08 frame. This can be used for the future to see how well different configurations may fit in an already designed frame.Other Universities is a folder of simple research done on last year’s hybrid teams. We did research on some of the top competitors; McGill, Illinois, Drexel, Yale, University of California, and Embry-Riddle Aeronautical University of Research. Each of these universities cars were broken done into drive motors, accumulators, engines, drive trains and other important information. For each of these categories you will find specs of each component as well as prices and weight in most cases. We did start are own road loading model in Simulink. This ended up helping us understand concepts but was found ineffective due to the fact that PSAT did almost everything we conspired to use this program for. There is also a presentation in this folder that may be useful for basic understating of our Simulink road load model.Track information is currently a small folder. It contains a picture of the 2008 FSAE track, which is very likely to change in the next couple of years to incorporate a more regenerative braking oriented course. In the Purchased folder you can find all of the specifications on each purchased part for the single and double motor test. This includes belts, pulleys, load cell, Hall Effect sensors, etc. The Minutes/Agendas folder is very clear-cut. In this folder you will find every Agenda along with minutes taken during meetings. Most of the minutes are taken during sponsor meetings where as other minutes and agendas can be found in each team member’s notebook.Project tracking is also a very straight forward folder. Here you will find the House of Quality and a time line for the first semester as well as a design process summary. This file makes it easy to see where the FSAE team has been as well as where we want to be. The Team Tool Box folder is something we use for any useful information from past teams or ordering information for components. All phone numbers for different companies for specific components will be stored here. Right now all that exists in this file is the phone numbers of last year’s team members.The Website folder contains everything that exists on the HYROLLERS internet page; which can be viewed at seniordesign.engr.uidaho.edu/2008_2009/hyrollers. Presentations & Papers is self explanatory. In this folder all revisions of final papers can be found as well as presentations along with posters prepared. All of the appendices can be found in the folder called appendices. The Resources folder is a great place to find information from either past teams or information gathered from last year’s completion. The folder names are as follows; hybrid rules/08’ program/& others, Hybrid 07-08, 07-08 Car, 06-07 Car. Also you will find helpful hints using the program PSAT, which is a program using MATLAB and Simulink to predict the outcomes of specified drive trains for hybrid vehicles “Power Systems Analysis Tool Kit.” Jason Sagen’s Thesis is also found in this folder, “Hybrid-Electric Planetary Drivetrain Design for Automotive.” This is a resource that our team found very useful in understanding the concepts of a hybrid vehicle. The PSAT folder is empty due to time. We have not yet explored this program enough to have a valuable understanding of what it can do. In the Hybrid rules/07’ Results/& Others folder you will find the 2008 Hybrid Rules as they are of now, a folder containing who competed in the 07’ program along with the results, and a breakdown of the U of I FSAE west competition. Hybrid 07-08 is a useful folder in the fact that it has all of the solid modeling of the Yamaha YZ250F engine that we will most likely be using for the final Hybrid vehicle. The 07-08 Car along with the 06-07 car folders is simply all of the useful information that our team could see helpful from previous years FSAE teams. These folders contain useful solid modeling and information on frames, suspension, steering, braking, etc. In the Pictures folder you will find a number of pictures of the phase 1 test “single motor.” There is also a number of other pictures of all the components used; such as the load cell, hall effect sensor, DC power supplies, and Dacu.The Single Motor Data folder two excel files are located. One contains all of the load cell calibration data along with the calibration curve. The other contains data taken from the load cell and hall effect sensors as well as the final torque curve in addition to a statistical analysis of all the data taken. Lastly, the Test Bed Modeling file contains the drawing package on top of all the solid modeling files created for the full test bed model for both the single motor test along with the duel motor test. Appendix MTotal ExpensesHyRollers' Budget Component DescriptionEstimated CostsActual CostsBase matl's$10.00$55.00$64.72$10.50Bearings$4.00$116.62Coupler$50.00$84.24Power Transmission (Pullys and belts)$100.00$153.45*Motor 1$110.00$98.99Controller 1$90.00$81.00Throttle control1$90.00$89.99Motor 2$110.00$109.99 Additional Controllers$250.00$238.002 Hall Effect sensors$50.00$66.75Computer-I/O interface$50.00$45.15Hardware$20.00$34.00Multi-Power supply$150.00Not needed2 Load Cells$100.00$176.00Unforeseen and Miscellaneous$250.50$91.07$1,500.00$1,449.97$50.03<--- Remaining Funds* Two additional belts will need to be purchased at an expected cost of $15.00 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download