5/6/2019 2019 Tech Report - Google Docs
University of Rochester Solar Splash
Boat #8 Technical Report
May 6th, 2019
Team Members Chris Dalke · Andrew Gutierrez · Seth Schaffer · Martin Barocas · Nick Davis Ivan Frantz · Leo Orsini · Owen Goettler · Daniel Wong · Heriniaina Rajaoberison Melanie Armenas · Mark Westman · Daniel Allara
Advisors Dr. Ethan Burnham-Fay · Kyle DeManincor · Jim Alkins
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 1/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 2
Executive Summary
This year, the University of Rochester Solar Splash team focused on improving our boat in two areas. We built a robust, modular telemetry system which records sensor data about the boat’s performance. Data points are displayed live in the boat, transmitted over radio to a shore computer, and stored in a database for later analysis. This system is used to transmit the throttle and controls for the boat, allowing our team to use control algorithms to follow a power budget. On the mechanical systems side, we improved our outboard motor to use a simplified direct drive and interchangeable motor system. Our outboard motor has been modified to accommodate propellers up to 14”. The ability to change motor and propeller on the same outboard assembly gives our team a large amount of flexibility. Each of these changes was implemented in response to events of the previous year. At the end of last year, our team suffered a setback that prevented us from attending competition. During testing, we overheated our motor and were not able to secure a replacement in time for competition. This incident, while unfortunate, was incredibly valuable in directing our team’s future. First, this was an indication that our drivetrain, propeller, and motor were mismatched as our system was overloading the motor even at low speeds. Second, this reinforced the importance of a telemetry system which can provide feedback early in the testing phase to identify problems. With these changes in mind, our team set goals for expanding our telemetry system and modifying the drivetrain. Our efforts this year are also a continuation of our goal for last year of building a modular system which can be iterated on yearly by our team. Although our focus this year was on two subsystems, we expanded the modular nature of our boat in both the telemetry and drivetrain systems. The end result of these efforts is a boat which can be reconfigured quickly, is easy to setup and run in testing, and is robust to the failure of an individual component. The development of a telemetry system is a huge step forwards for our team as it allows us to quantitatively measure the results of our system improvements. Previously, we had no system of measuring the performance of our boat, meaning we could only guess at where problems in our system existed. The telemetry system allows us to accurately measure and compare the performance of our boat, as well as diagnose problems in a more efficient manner. Outside of engineering work, our team has spent time recruiting members, participating in on-campus and off-campus events, and other matters which will ensure the longevity of the group. We implemented training programs in electronics, programming, and mechanical design, with the intention of helping members build engineering skills. Overall, we are very proud of the work that the University of Rochester Solar Splash team has done this year. We’ve made significant progress in our engineering work, involvement in our campus community, and overall mission of racing a successful boat at the Solar Splash competition. We look forward to competition!
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 2/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 3
Table of Contents
Executive Summary 2
Table of Contents 3
I. Overall Project Objectives 6
II. Data Acquisition and/or Communication 7 A. Current Design 7 B. Analysis of Design Concepts 7 C. Design Testing and Evaluation 8
III. Electrical System 9 A. Current Design 9 B. Analysis of Design Concepts 9 C. Design Testing and Evaluation 11
IV. Power Electronics System 11 A. Sprint Configuration 11 B. Endurance Configuration 12 C. Power Budget 13
V. Drivetrain and Steering 14 A. Drivetrain 14 B. Propellers 16 C. Steering System 17
VI. Solar Energy System 19 A. Solar Panels 19 B. Solar Chargers 19 C. Solar Tracking System 21
VII. Hull Design 22 A. Current Design 22 B. Analysis of Design Concepts 22 C. Design Testing and Evaluation 23
VIII. Project Management 23 A. Team Members and Leadership Roles 23 B. Project Planning and Schedule 24 C. Financing and Fundraising 24
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 3/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 4
D. Strategy for Team Continuity and Sustainability 25 E. Discussion and Self-Evaluation 25
IX. Conclusions and Recommendations 25 A. Strengths 25 B. Weaknesses 26 C. Lessons Learned 26 D. Achievement of Objectives 26 E. Future Objectives 26
References 27
Appendix A: Battery Documentation 28 A. Material Safety Data Sheets 28 B. Battery Datasheets 34
Appendix B: Flotation Calculations 36
Appendix C: Proof of Insurance 38
Appendix D: Team Roster 39 A. Undergraduate Roster 39 B. Advisor Roster 39
Appendix E: Telemetry Data 40 A. Description of Telemetry Protocol 40 B. Telemetry Node Implementation 42 B. Telemetry Node Data 43
Appendix F: Electrical System Data 44
Appendix G: Power Electronics Data 47 A. Motor Specs 47 B. Motor Controllers 47 C. Motor Datasheets 48 D. Notes on Alltrax Throttle Control 51
Appendix H: Drivetrain and Steering Data 52 A. Torqeedo Propeller Datasheets 52 B. Drivetrain CAD Drawings 53 C. Lower Unit Modifications 58
Appendix I: Solar Energy System Data 59
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 4/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 5
A. Solar Charger Datasheets 59 B. Solar Panel Datasheets 60 C. Explanation of Step-Down vs. Boost MPPT Chargers 61
Appendix J: Hull Design Data 63 A. Hull Modification Process 63 B. Hull Design History 67 C. Hydrostatics Testing Program 69
Appendix K: Project Management Data 71 A. Executive Board Roles 71 B. Budgeting Documentation 71
Appendix L: Sponsorship Packet 73
Appendix M: Power Budget 78
Appendix N: Solar Tracker 80
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 5/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 6
I. Overall Project Objectives
Our objectives for this year focused on two areas: Our drivetrain and telemetry systems. We identified several objectives in each area, which we worked on throughout the year. In the drivetrain area, we had the following goals: ● Move to a direct drive system: Previously, we were gearing up our outboard motor to attempt to raise the propeller RPM, but this overloaded the motor since it was not able to provide the torque requirement. Instead of using the mechanical system to force the desired output RPM, we planned to use a direct drive and a motor which would better fit for our application without gearing. ● Build a system for switching motors on the assembly: The motor we use for Sprint is not efficient for Endurance. In order to pair a motor to each event, we need a system which allows us to easily interchange motors on the same assembly. ● Adapt the lower unit to allow for larger propellers: The outboard motor unit that we use previously only allowed for 8” propellers due to a ventilation plate. By removing this plate, we can accommodate propellers up to 14”. This allows our team to use a wider range of propellers, particularly larger diameter propellers which operate at lower speeds that better match our electric motors compared to the original boat propellers. In the telemetry area, we built on the work we started last year: ● Instrument all pieces of our electrical & drivetrain system: We set a goal to complete the work started in previous years, and build telemetry sensors for all components of our system. This involved the creation of several telemetry sensor PCBs. ● Complete the Telemetry functionality: The telemetry system needed a completed monitoring UI, database storage, and a more robust communication protocol which could handle the increased sensor input. ● Shift throttle and control signals to use the Telemetry system: Previously, our throttle used an analog voltage connected directly to the motor controller. We are switching to a throttle value within the telemetry system, which is interfaced with the motor controllers via a specialized telemetry device. This consolidates our signals to use a common system, and allows us to control throttle more accurately. ● Use sensor data to aid driver decisions in endurance mode: In the endurance race, the goal is to target a set power consumption as governed by our power budget. Once we have moved to a drive-by-wire system, we can use a PID loop in the telemetry system to drive the boat at a fixed power consumption. In summary, our team has objectives in two areas: Improvement of our drivetrain system, and the completion of our telemetry system. The pursuit of these goals has allowed our team to develop a system which can be modified in response to testing and provides extensive data about performance, setting our team up for success at competition and in future years.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 6/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 7
II. Data Acquisition and/or Communication A. Current Design Our previous data collection system used a Raspberry Pi, which connected to a battery monitor to collect voltage and current readings using a JSON-based protocol. The data was collected and stored as a .csv with each row representing a second of data. The system also sent data over radio to a shore computer, also via JSON format. Our team built a python-based UI which displayed the data.
Right: Last year’s telemetry UI, which displayed a simple chart of voltage and amperage over time.
B. Analysis of Design Concepts This system was a good start, but had minimal features. This protocol was slow to collect data because each datapoint contained text making the packets extremely large. There were issues with buffers overflowing, lag in receiving the data points and logging the data. In addition, there was no way to manage multiple telemetry devices and expand the system easily. The only way to save the data was to dump it to a file from the python utility, which would freeze if there were too many data points. This year, the Data Acquisition and Communications system (Referred to by our team as the “Telemetry” system) is at the center of our boat’s strength. The system design was expanded greatly, and is now used for all boat control signals. The new telemetry system consists of several parts, each of which have been developed this year: ● Telemetry Server: A central hub which aggregates sensor values, stores and transmits these values, and runs algorithms such as a throttle feedback control loop. ● Telemetry Protocol: The communication method(s) used to pass data between the telemetry server and sensors. ● Telemetry Sensor Nodes: Computational nodes that collect data from related sensors and transmit packed binary data to telemetry server. ● Telemetry Monitoring UI: Software, both on the boat and onshore, for monitoring and analysis of the recorded data.
The Raspberry Pi-based design from the previous year has been expanded, and the Pi is now the center of the boat’s systems. The “Telemetry Server” is responsible for managing connections over USB to devices. The server constantly scans USB ports for devices and manages connecting to sensor nodes around the boat in addition to decoding, aggregating and interpreting all the data sent to it. In addition, the server is set up with a 433Mhz radio transceiver used to stream all data back to the shore for remote analysis.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 7/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 8
The Telemetry Server unit, which serves as the hub for all telemetry devices in the boat. ① Waterproof USB plugs ② Power Input ③ HDMI Output ④ 433Mhz Radio Antenna ⑤ Power Switch ⑥ Status LED
The server uses a protocol designed by our team, the “Telemetry Protocol”, to communicate with sensor nodes. This binary protocol replaces previous text-based protocols. The system has been designed to be extremely lightweight, allowing for transmission of a large amount of sensor data in minimal space. For more information about the Telemetry Protocol, please see Appendix E: Telemetry Data. To better view and analyze this data, we have updated our Telemetry UI software. The new software displays live data from all the data points, as well as charts for the most important data points. The system runs on the boat dashboard as well as on a shore computer. To obtain data, the UI requires a telemetry server instance running on the same computer it is running on. The telemetry server system uses the Telemetry Protocol to maintain a connection with the boat over radio.
Right: The updated Telemetry UI interface, with charts view showing data such as voltage and current.
C. Design Testing and Evaluation There are several areas for future improvement, evaluated based on our testing. There are several data points we do not currently receive, such as battery terminal voltage and individual cell voltage. A telemetry node should be built to track these data points. Additionally, the dashboard UI on the boat is laggy since the Raspberry Pi cannot handle the load of rendering the UI smoothly. The boat UI should be simplified to just show a live view of the most important data points, which will improve performance. This modification may be completed before competition.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 8/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 9
III. Electrical System A. Current Design Last year, our electrical system consisted of a battery box, throttle box, motor controller, solar controller, telemetry box, and bilge unit. The battery box used a Victron BMV-712 monitor for voltage and current monitoring capabilities. The bilge unit housed an auxiliary battery to power the bilge pump. All other components of the electrical system have been discussed in their respective sections.
Right: Victron Energy BMV-712 monitor, and 500A shunt resistor used in the battery systems.
B. Analysis of Design Concepts We have kept the same battery box system. The BMV-712 unit communicates with the telemetry server over USB, and its data is integrated into the telemetry system. Since we moved to a drive-by-wire throttle, the original throttle box was removed and replaced with a telemetry node which takes a throttle input and sends it to the telemetry system. As much of the team’s focus this year was on collecting data to evaluate subsystems and guide design decisions, there were many new additions to the electronics systems to interface with new peripherals to obtain this data. Several sensors were added, discussed below. Schematics for all telemetry devices can be found in Appendix F: Electrical System Data. 1) Motor Board: This sensor collects motor speed and temperature data. The board reads a hall switch used to calculate motor and prop RPM. There is a collar attached to the shaft of the motor that houses a small magnet aligned so it will trigger the hall switch once per revolution. This signal triggers an Interrupt Service Routine (ISR) that keeps track of the number of revolutions within a time window. The motor board also calculates sprint motor temperature by reading an embedded thermistor. The voltage drop from the changes in resistance are measured with a voltage divider read by a ATMega32u4 based microcontroller. This reading is then converted to a temperature on the microcontroller based on calibration data.
CAD model of the RPM sensor assembly used with the motor board.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 9/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 10
2) Motor Controller Board: The motor controller board (Shown, right) is used to write throttle signals to the sprint motor controller and read data from the internal registers of the controller through the RS-232 serial port. The motor controller board has a socket for the microcontroller which handles all of the computation and interfacing with the telemetry protocol. There is an MCP-1416 digital potentiometer on the board wired to the throttle input of the sprint motor controller used by the microcontroller to write the throttle value sent by the server. To read the data stored by the controller, the node has a UART to RS-232 converter setup with an implementation of the Alltrax AXE protocol used by Controller Pro. The binary data from the controller is read and processed on the node before it is sent back to the server. There is also an output for a 5V relay which is used to control power to the sprint motor controller. All these sensors are packaged on a custom PCB designed using easyEDA with a common-cathode RGB indicator used to show the status of the controller. 3) Solar Charger Node: The solar charger node (Shown, right) is used to determine the output current of the solar chargers to the batteries. This is used to evaluate how much energy the system is receiving from the sun and forward it to the server to budget energy for use in the endurance competition. There is a socket for the main microcontroller and two hall-effect based, bi-directional 50A current, one for each of the solar chargers. These sensors were also packaged into a custom PCB with traces sized to handle a maximum to 14 A each with a 20℃ rise. This is so that each of the two sensors can independently handle all the current from our solar panel while still managing the heat transfer appropriately. The solar node is mounted in the a box with the solar chargers and setup to take input from the solar panels. More information on tha solar energy system can be found in Section VI: Solar Energy System. 4) Throttle Board: The driver uses a throttle knob to control the speed of the boat. The throttle board connects to a potentiometer, reading the value and sending it to the telemetry system. This board is very simple, consisting of a connection to the potentiometer, a pull-down resistor in case the potentiometer is disconnected, and a status LED. 5) GPS/IMU Board: This board uses a GPS chip and accelerometer/gyroscope to record the speed and orientation of the boat at all times. The Arduino communicates over serial to an Adafruit Ultimate GPS Breakout and an Adafruit 9-DOF IMU board. The code filters and
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 10/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 11
processes the data, taking a rolling average of the accelerometer data to eliminate noise. Finally, this data is sent through the telemetry system.
C. Design Testing and Evaluation Each of these boards was constructed as a prototype using a breadboard. Once we were satisfied with the functionality, we designed and ordered a PCB. The PCBs were assembled, tested, and integrated into the boat systems. Future efforts will include expanding the telemetry sensors to include more nodes. Because the system is modular, each individual PCB collects only a few data points. Adding new boards which collect more data is designed to be convenient under this system.
IV. Power Electronics System A. Sprint Configuration 1) Current Design: Our previous competition entry used a single brushed motor for both Sprint and Endurance modes, the Lynch LEM 170-95 Permanent Magnet DC motor. This motor, purchased during the 2017 year, is capable of a peak power rating of 16 kW and continuous rating of 7.2 kW. The motor was connected to a belt-drive system which transferred the power to a modified outboard motor, discussed in the Drivetrain and Steering section. The motor was driven by an Alltrax AXE 4844, a 24-48v brushed motor controller. We controlled the AXE via a 0-5v potentiometer signal sent over an ethernet wire to a throttle unit. For batteries, we utilized a set of nine Enermax G13EP 13 Ah batteries in a 3-series 3-parallel configuration, giving us a total system equivalent of 39 Ah at 36 volts. This gave us the benefit of a high peak discharge current with lower equivalent terminal resistance and load on individual batteries as compared to a 3-series system of larger batteries.
The battery layout utilized in last year’s sprint mode, with 3 series-connected sets of 3 parallel batteries, for a total of 39Ah at 36v.
2) Analysis of Design Concepts: As mentioned in the Executive Summary, last year we overheated the LEM 170-95 motor during testing. Our system was drawing excessive current from the motor, and we did not have a monitoring system in place for motor temperature. The failure occurred in the brush holder assembly, where the brushes of the motor heated up to a point where they melted their plastic guides. Right: The brush holder assembly with failure visible on the right brushes.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 11/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 12
After evaluating the reasons for this failure, we concluded that a number of errors in the mechanical and electrical design of our system caused this failure. We improved the drivetrain to prevent such an overload of the motor, and our telemetry system now collects temperature from the motor to provide driver feedback and prevent overheating. When replacing the motor, we purchased the larger LEM-200 95. This motor has a higher rated torque output and should be able to better handle the power requirements of our application.
LEM-200 95 brushed motor used by our team in Sprint mode. The motor is shown mounted on its adapter plate for the direct-drive system used by our team this year.
Despite the motor failure, the motor controller and other power electronics are still satisfactory. We chose to continue using the Alltrax motor controller with the new motor. With the development of our telemetry system, we are using the telemetry network to transmit a digital throttle value. With the new system design, the throttle is sent as a data point into the telemetry system, and is then transmitted to the motor controllers over a serial connection. Since the Alltrax cannot be controlled directly via serial data, we needed to create a circuit which would convert the throttle control signal to one of the inputs of the Alltrax controller. We built a sensor node which converts the telemetry data point to a resistance value, which the Alltrax controller reads and converts to a throttle value. (For more information, see Appendix G, Section D: Notes on Alltrax Throttle Control.)
3) Design Testing and Evaluation: We performed testing throughout the year. In the fall, we tested extensively with the outboard motor on a testing rig which allowed us to submerge the propeller in a tub of water. While this did not fully approximate the load of a moving boat, it allowed us to test the drivetrain. We tested our boat on the water in December, which validated our design choices. Last year our boat travelled a maximum speed of 8.7 mph with a maximum current draw of approximately 300A. During the december testing session, we reached a speed of 9 mph with a current draw of approximately 80A. This represents a significant increase in efficiency.
B. Endurance Configuration 1) Current Design: As mentioned above, our team previously used the same brushed motor for both Sprint and Endurance modes. Our endurance battery pack consisted of three Genesis EP 42 Ah batteries in series for a single 42 Ah, 36v battery set.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 12/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 13
The battery layout for last year’s endurance mode, consisting of three 42 Ah batteries in series for a total of 42 Ah at 36v.
2) Analysis of Design Concepts: This system was inefficient for endurance due to the high no-load current consumption of the Lynch motor. Our team decided to move to a smaller brushless DC motor for endurance mode. We chose the Alien Power Systems 6374S motor, which is a 42v, 2.8kW rated motor. We chose this motor because its speed constant is similar to the rated speed for the propeller we plan to use for endurance, and because it has been used commonly with the VESC motor controller.
Left: The Alien Power Systems 6374S motor used by our team for Endurance mode. Right: The VESC 4 motor controller used in Endurance with the motor.
To drive the BLDC, we chose the VESC 4 motor controller, shown above, an open source motor controller originally developed for usage with electric longboards[2] . The VESC is popular for its heavily customizable firmware and extensive sensor data. The VESC can be controlled through a USB serial connection, and the telemetry server implements the VESC protocol, allowing it to communicate with the device. This is extremely convenient because it allows us to control the VESC and track data points without an intermediary microcontroller as required with the Alltrax. To interface our Python telemetry program with the VESC, we used the PyVESC library[3] . 3) Design Testing and Evaluation: Similarly to the Sprint power electronics, we performed testing in a variety of conditions, including no-load bench testing, testing on a rig, and on-the-water testing. We have not performed speed testing with the endurance configuration yet, but plan to continue testing up to competition.
C. Power Budget Discussed in the Overall Project Objectives section of this report, one of our primary technical goals was to create a power budget that would be enforced during endurance to ensure that our boat is able to keep a constant speed during the entire endurance heat without being required to stop.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 13/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 14
The formula and procedure for the power budget is found in Appendix M: Power Budget. Our calculations found that our team should aim to draw approximately 20 amps, assuming a 60% solar efficiency. By designing for this specification and driving the boat to not exceed this value, our team will be able to successfully complete endurance without stopping the boat. In future competitions, we plan to build the power budget into a program running on the boat. This program will automatically adjust the maximum throttle based on the energy received from the solar panels and the state of the battery system. This automation will further help optimize our system for endurance.
V. Drivetrain and Steering A. Drivetrain 1) Current Design: Our previous design consisted of an Evinrude outboard motor with the gasoline upper unit replaced with an electric drive unit designed by our team. The drivetrain used a 1.78:1 belt drive reduction which connected the electric motor to the outboard shaft. The intention of the belt drive was to increase the output RPM enough to allow the team to use conventional propellers.
Left: CAD model of the latest revision of the belt-drive upper unit. Right: An early version of the outboard mounted on a boat during testing.
The original Evinrude outboard motor had several constraints which our team attempted to overcome using the design of the upper unit. First, the lower unit has a gear reduction of 1.87:1. Additionally, the original Evinrude lower unit would only allow for propellers up to 8” diameter. 2) Analysis of Design Concepts: There were several issues with the drivetrain unit that we aimed to fix in this year’s iteration. The core of our issues was due to the belief that the motor could provide enough torque while geared up nearly 2:1. The intention of the gearing was to increase the prop speed to generate more thrust. However, the motor could not generate enough torque, meaning the system pulled more current than the power electronics and batteries could accommodate. In addition to the electrical issues associated with the gearing, it was outputting at a decreased power because the system was operating outside the intended operating range of the motor. In order to get the best possible performance out of the motor, we should not rely on overloading the motor to lower the RPM, instead we should drive the motor at the maximum RPM while it still has enough torque.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 14/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 15
Based on this, we planned to use a direct drive, and spend time identifying propellers that will match the characteristics of the motor we use. Since this strategy would require a wider range of propellers, we planned to modify the Evinrude lower unit to accommodate larger propellers, up to 14”. We designed a direct drive system which connects the motor through a jaw coupling to the outboard shaft. The direct drive unit has been designed to make it easy to switch between Endurance and Sprint motor. Each motor is mounted on its own plate, and the plates can be interchangeably bolted down to the upper unit. Each motor has its own upper jaw coupling which stays mounted at all times to the motor and meshes with the bottom unit. For detailed CAD drawings of the components, see Appendix H, Section B: Drivetrain CAD Drawings.
Left: The upper drive unit shown with the Sprint motor installed. Right: The upper drive unit shown with the Endurance motor installed. Note the bolts on the upper plates, which are unscrewed to change which motor is used in the assembly.
An exploded view of the main drivetrain components in the upper unit; left to right: ① Motor, ② Upper jaw coupling, ③ Rubber “spider”, ④ Lower jaw coupling and shaft adapter, ⑤ Bearing, ⑥ Outboard shaft
We modified the lower unit by removing the ventilation plate, allowing for larger propellers. For more information about this process, see Appendix H, Section C: Lower Unit Modifications.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 15/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 16
3) Design Testing and Evaluation: After completing fabrication of the drivetrain, we performed several stages of testing. Early in the development, we performed bench testing. We built a test stand which submerges the propeller in a tub of water, allowing us to approximate the load of a boat in the water. Once the drivetrain system was completed, we tested on the water.
Left: The test rig, used for lab testing of the outboard. Right: The boat in the water during a testing session in December, where we tested the drivetrain’s efficiency at low speeds.
This testing validated our decision to update to a direct drive; the new direct drive system eliminates the majority of vibration and noise the drivetrain had previously, indicating it has increased mechanical efficiency. Additionally, our telemetry system allowed us to quantitatively compare the systems. On our old system, we had a maximum speed of 8.7 mph with a maximum current draw of approximately 300A. During the december testing session, we reached a speed of 9 mph with a current draw of approximately 80A. This represents a significant efficiency increase for our team. We have not yet tested the full speed of the boat, which requires a new propeller with a higher pitch.
B. Propellers 1) Current Design: Previously, our team used 8” propellers which were designed for use with the original Evinrude outboard motor. We were restricted in our propeller size by the ventilation plate on the outboard motor, and by the shear pin style of the lower unit. 2) Analysis of Design Concepts: The propellers that were designed for the original gas outboard were not suitable for our system. We evaluated many propeller options, before deciding on two Torqeedo propellers. We chose these propellers because Torqeedo motors have a very similar shaft RPM to our system, with a maximum RPM of 1600 for most models. Additionally, the propellers use a shear-pin style which matches our outboard.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 16/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 17
Left: The Torqeedo v8/p350 propeller we plan to use for Endurance competition events. Right: The Torqeedo v30/p4000 propeller we plan to use for Sprint competition events.
Data about the performance characteristics can be found in Appendix H, Section A: Torqeedo Propeller Datasheets. The Sprint propeller is rated for 30 knots, and the Endurance propeller is rated for 8 knots. These match our target speed ranges in each of the events. The diameter of the largest propeller, 12.6”, fits on our outboard with the modifications that have been made this year. 3) Design Testing and Evaluation: We have tested the Endurance Torqeedo propeller, and were able to achieve its rated speed of 8 knots. We consumed significantly more power with our Sprint motor than the original Torqeedo rating, which is expected because the Torqeedo system is more efficient. We have not yet tested the Sprint Torqeedo propeller. We plan to perform testing through the end of the semester in preparation for competition.
C. Steering System 1) Current Design: Our team has been utilizing electrical steering systems for several years, including two years ago, when we last attended competition. This was motivated by the presence of the custom drivetrain / hull that made it difficult to use a conventional steering system. This steering system used a linear actuator and lever to steer the motor back and forth, as shown in the diagram to the right.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 17/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 18
2) Analysis of Design Concepts: This electrical steering system was very ineffective. The design of the lever meant that large off-axis forces resulted in high bearing stress on the bolts causing them to shear during competition. Second, the steering system had no feedback through resistance on the wheel, so it was very hard to control and several times during competition our skipper lost control of the craft. Additionally, there was an extra degree of freedom in the lever system used which made off-axis alignment possible, jamming the steering. After analyzing the options for the steering system, the team elected to return to a mechanical Teleflex rack-and-pinion steering system. This matched up well with the decision to use an older hull, because the hull had mounting holes for a Teleflex adapter used previously.
3) Design Testing and Evaluation: Because the model of our outboard did not have mounting holes for the teleflex cable directly, we mounted the Teleflex rear unit to the transom of the boat underneath the motor. This required the team to manufacture a rod to connect the end of the steering cable to the outboard. This connecting rod was manufactured from a ¾” steel rod, with ball joints on either end to allow for rotation. The layout of the steering system is shown in the figure to the right. The Teleflex cable was attached to the side rail of the boat, and the steering unit was mounted on the dashboard. To minimize moments on the bolt, we used a shorter bolt and connected the steering as close to the height of the outboard mount as possible. We tested this design extensively during on-the-water testing, and it performs as expected. Because the Teleflex system provides feedback to the driver through the force required to turn the steering wheel, this system is much easier to steer than the previous electronic steering system.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 18/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 19
VI. Solar Energy System A. Solar Panels 1) Current Design: We previously used two panels manufactured by SBM Solar, Inc. Datasheets for our solar panels can be found in Appendix I, Section B: Solar Panel Datasheets. Each panel is rated to output 258W, bringing our total wattage to 516W which is above our limit of 480W for premade panels. To reduce the panels to the legal limit per the rules, we covered a proportional area of the panels using masking tape. This solution, suggested by the judges in previous years, dropped the rated output power to within the legal range. To calculate the area to be covered, we used the following calculation:
Allowed W attage 480 Legal Area = Our W attage = (258 × 2) = 0.9302 = 9 3.02% of Solar P anel P ercent to be Covered = 100% − Legal Area = 100%− 9 3.02% = 6 .98% (7%) of Solar P anel Area to be Covered = P ercent to be Covered × Solar P anel Area Area to be Covered = 0.0698 × ( 2515.87 in2) = 1 75.61 in2
Based on this calculation, our team covered approximately 175 square inches of the panel. To mount the panels, we previously used an aluminum Z-channel frame around each panel. Holes were drilled in the outer edge of the aluminum channel for bolts and tie-downs.
2) Analysis of Design Concepts: Our solar panels are satisfactory, and we have no current plans to upgrade them. We made minor modifications to the mounting system for the panels this year. The previous system had numerous dents and holes from years of usage, and it was inconvenient to attach and detach from the boat. To improve the design, we designed a new frame made from aluminum square tubing, with threaded holes on each corner. The solar panels are attached to the boat hull or solar tracker through the use of threaded knobs, which are easier to use in the field than bolts.
Right: Mounting system for the solar panels. ① Threaded Knob ② L-Bracket Mount (on Boat) ③ Frame Corner ④ Square Tubing
3) Design Testing and Evaluation: At this time, we are fabricating the new solar panel mounting system. We plan to evaluate the usability of the new system during testing and at competition. B. Solar Chargers 1) Current Design: In our previous design, each solar panel was connected to a Genasun GV-Boost 8A 36V charger, with both charge controllers connected in parallel to the battery system through a shunt resistor for current monitoring purposes, as seen in the schematic below.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 19/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 20
The schematic for the parallel-wired Genasun controllers used in our team’s previous design.
These chargers were chosen for their voltage step-up capability, allowing the use of solar panels at a lower voltage than the batteries. Most MPPT solar chargers are step-down converters, meaning the solar panel voltage must be higher than the battery voltage. A boost converter is required in our system to convert the lower input voltage from the panels to a higher voltage for battery charging. (For an explanation of the background behind this, see Appendix I, Section C: Explanation of Step-Down vs. Boost MPPT Chargers.) The entire system was contained in a waterproof enclosure with quick-disconnect connectors to the solar panels and battery.
Right: The previously completed solar charger module, with solar charger input / battery output cables, current monitoring LCDs, and an on/off switch.
2) Analysis of Design Concepts: The solar chargers we chose are effective for our purposes, and we have no plans to replace them. We have designed improvements to the monitoring capability of the solar system, to fit in with our larger telemetry system efforts. Although the previous solar charger system had current monitoring capability, the value was only displayed physically on an LCD, meaning it was impractical for the driver to read while in the boat. Additionally, the data was not saved for later review. We designed a sensor board with two separate shunt resistors, which measure the output from each individual solar charger. The solar charging current is passed through a PCB which reads the charging current and sends this data to the telemetry system. We verified that the PCB traces could support the maximum charging current with acceptable power loss using an online trace width calculator[1] . Another concern with the previous solar system was the size of the enclosure. We purchased a smaller enclosure and used a 3d printed mounting system on the interior of the box to mount the solar chargers and telemetry board. To further improve the size of the components, we began using XT-60 and XT-90 connectors in place of the larger Anderson power connectors used previously.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 20/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 21
Updated enclosure for the Solar Energy System. ① Telemetry USB connection ② 36v battery output ③ Solar panel inputs
3) Design Testing and Evaluation To test our solar system, we repeatedly brought the panels out in a variety of sunlight conditions. During these tests, we connected a dummy load in the form of a motor running at full throttle which drew approximately 10A (360w). While this testing was not qualitative, it allowed us to debug potential problems with the telemetry sensor data, and we verified that the charging system works as intended. As we approach competition, we plan to perform endurance testing on the water with the new system.
An example of our solar testing in the field: Both panels connected to the system in dusk & partial shaded conditions. Our team plans to perform a number of these tests at different light intensities and weather conditions.
C. Solar Tracking System We constructed a solar tracking system for this year’s competition entry. Because this is not a system-critical project, we have placed a description of the project in Appendix N: Solar Tracker.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 21/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 22
VII. Hull Design A. Current Design In previous years, our team has produced several self-built hull designs. The team has produced several iterations of fiberglass hulls, and the most recent design constructed by our team was a hand-built cedar strip boat hull. A brief history of hull designs is shown in Appendix J, Section B: Hull Design History. Last year, our team returned to a premade hull which was purchased at the inception of the club. This hull, a 16-ft square-stern cargo canoe, performed well in at the 2010 competition. By using the preexisting hull, our team was able to narrow our focus on other subsystems with the interest of improving a single area of our boat’s design at a time. Pictured Right: The Ungava Bay 16 flat-stern canoe that we planned to utilize if we had attended last competition year. We have used this boat extensively for testing.
B. Analysis of Design Concepts This year, we elected to continue using the square-stern canoe, under the same philosophy that spending less time on hull design would allow our team to focus its limited efforts elsewhere. After meeting project milestones for the drivetrain and telemetry systems, we were left with time to work on modifications to the existing hull to ensure it was competition ready. We had the following concerns: - The transom, benches, and buoyancy chambers were made of marine plywood, and had rotted out over the 10 years the team has owned the hull. This increased the weight of the hull and meant that the buoyancy was compromised. - The benches were not in ideal locations; their positioning made it hard to fit the driver, electronics, and solar panels. - The front of the boat had a canoe-style bow which curved upwards. This made it very difficult to mount a solar panel towards the bow of the boat, further restricting our available space. To solve these issues, we planned a hull reinforcement design and set of modifications which would improve the hull’s usability at competition.
Left: Closeup of the transom, which had rotted out and needed replacement. Right: CAD model of the planned hull reinforcements which would replace the benches.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 22/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 23
C. Design Testing and Evaluation We performed the hull modifications in the winter, when on-the-water testing was impossible. We started by stripping the hull down to its fiberglass shell, removing stringers, transom, and gunwale. Next, we fabricated and reinforced the parts for the frame. We purchased 16’ segments of white ash, allowing us to use a continuous piece of wood to reconstruct the gunwale. Before attaching the reinforcements to the hull with fiberglass, we encapsulated all wooden parts with epoxy. After assembling the reinforcements, we painted the interior with marine top coat paint. The buoyancy boxes were filled with closed-cell floatation foam. We built a deck using HDPE plastic material (“King Starboard”) used commonly on boats; this provides a waterproof surface covering the floatation foam. After completing the structural components, we added a system of straps and tie-downs which are used to mount all the electronics in the boat hull. The hull is now completed and ready for testing before competition. For a more detailed timeline of the hull modification process, see Appendix J, Section A: Hull Modification Process.
Pictures of the modified hull, ready for water testing and competition.
Separately, our team spent time learning hull design fundamentals. We constructed prototype hulls in Solidworks, and members trained to use ANSYS Fluent to run fluid simulations on hull models. To improve this process, we built a tool, Float-Util, which can be used to test the hydrostatic properties of a hull very quickly without running a full fluid simulation. Details of this tool and its usage in hull modelling can be found in Appendix J, Section C: Hydrostatics Testing Program. We plan to construct our own hull in the future, but our team must first gain a solid understanding of the fundamentals of naval architecture before moving forward with a full hull project.
VIII. Project Management
A. Team Members and Leadership Roles The University of Rochester Solar Splash team is a club team consisting solely of undergraduate students. The club is led by an executive board, which is an elected panel of leaders who are in charge of leading the overall direction of the project, as well as holding weekly meetings with advisors and faculty. The executive board also spends time keeping up
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 23/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 24
communication among the team and ensuring the team stays on schedule. For a detailed description of the roles in the executive board, see Appendix K: Project Management Data.
B. Project Planning and Schedule Project planning was divided into several categories by subsystem of the boat: Electrical System, Mechanical System, Telemetry System, Drivetrain, Hull Design & Simulation, and Construction & Manufacturing. Our team set goals for each of the design areas, and kept track of our progress towards each of our goals. Our planning mirrored the modular nature of our system; each subsystem had a separate task board maintained and the group responsible for that system kept the list updated. In previous years, our planning was relatively informal, and we did not set week-to-week deadlines. We previously used Trello as a management system, which provided a TODO list but did not have the ability to track progress of a larger project in depth. This year, we transitioned to Asana, a work management platform with a more advanced feature set as compared to Trello. We received a sponsorship for a premium Asana workspace. One of these features, Timeline view, was essential to our team’s project planning. Timeline view which allowed our team to organize tasks in a calendar view with other tasks as dependencies. We also used GroupMe, a group communication software, extensively to organize meetings.
An example of the Asana Timeline view, which allowed our team to build a detailed map of tasks and their dependencies in a calendar view. This allowed our team to keep on schedule throughout the year, and respond when plans changed.
C. Financing and Fundraising The Rochester Solar Splash Team is funded annually via two sources: The Hajim Engineering School and the Rochester Student Association. As an engineering club, our group is responsible for submitting a yearly budget to both institutions to continue receiving funding. The Hajim Engineering School contributed $6,500, and the Student Association contributed $3,870 for a total budget of $10,370. This budget covers our engineering expenses, as well as the travel expenses including hotel and rental truck for the yearly competition trip. The team kept track of its spending using a set of spreadsheets that tabulated all the costs over the semester, which are shown in Appendix K, Section B: Budgeting Documentation.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 24/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 25
The team earned supplemental funds through the recycling of old batteries and scrap metal which was previously stored in our shop and was due to be disposed of. Our team collected nearly 1200 lbs of lead acid batteries and 400 lbs of scrap metal and wire and brought it to a recycling center in Rochester, NY, where we received approximately $900[5] . This year, we set up a formal sponsorship program and created a sponsorship packet which has been a great success, helping our team obtain several material and software sponsors, such as Asana, a project management platform our team is now using for all task tracking. The sponsorship packet can be seen in Appendix L: Sponsorship Packet.
D. Strategy for Team Continuity and Sustainability Because we are an undergraduate club, our team relies on recruitment of new members to sustain itself. The primary method for recruiting new members is through presence at a variety of school events and activity fairs. Our team has achieved varied success with recruitment at these fairs in the past. Last year, we set a goal of becoming more active on the campus community. This year, we fulfilled that goal, participating in a number of events both on and off-campus. We participated in the ASME pumpkin launch, which is a competition of pumpkin launchers among engineering groups from our school. We placed 3rd at this competition. We also helped plan social events for the engineering school. Off-campus, we showcased our boat technology at the Rochester Maker Faire, a local festival with 3600+ attendees showcasing an intersection of science, technology, arts, and crafts.
E. Discussion and Self-Evaluation Overall, this has been a very productive year for our team. We achieved all the goals that we set in the beginning of the year, and are ending the year with a set of modular systems which can be expanded in the future. An ongoing concern is the future of the club’s membership; the majority of the work on this project has been done by a few individuals on the E-Board. In the future we would like to find a way to get other members of the club more involved, especially later in the semester. In the current situation, the members of the E-Board spend most of the time working on projects, and many general members stop attending meetings or working at the end of the semester when their workload increases elsewhere. Our team needs to develop methods that encourage these members to stay throughout the year. A priority for the veterans on the team needs to be recruitment and training for new members.
IX. Conclusions and Recommendations A. Strengths ● Our telemetry system is our biggest strength; the system allows our team to quantitatively evaluate the performance of the boat. The boat can actively respond to sensor data, preventing overload to electrical components in the system. ● The new direct-drive system is more efficient than our previous system, and can accommodate multiple motors and propellers. This has improved our speed and power consumption, and allows us to change motors quickly for endurance and sprint events.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 25/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 26
● The creation of a power budget will allow our team to perform better in the endurance section of competition. The Telemetry system includes a control algorithm which the driver can use to accurately target a goal power consumption during the event. B. Weaknesses ● We have not performed extensive testing of the maximum speed of our hull in Sprint mode. There will likely be issues that we have not anticipated with the system at high loads. ● Our hull is heavier than desired; the hull weighs approximately 200lbs. In the future, this is an area for improvement, as any weight reduction will improve our performance. There is also room for improvement in the hull form to achieve better performance. ● Membership numbers are an ongoing concern. This year, many of our team leaders who have been involved for multiple years are graduating. This leaves the team particularly vulnerable without leadership to push the group forwards. Additionally, There is a large amount of deep technical knowledge required to expand on the system that we have created. This makes it hard for new members to get involved. C. Lessons Learned ● In future years, the team should place a focus on involving more members in the central projects on the boat. Many of the newer members went through a training program and worked on smaller projects, but were not involved with the core boat systems such as the telemetry and power electronics. ● Future teams should start working on documentation of the systems as they build them; this helps pass along the knowledge so the system can be used later, and makes writing the technical report easier. D. Achievement of Objectives We achieved our goals for this semester. We were able to finish the new drivetrain design and telemetry system, and all of the goals identified in our Overall Project Objectives were met for this year. We are optimistic that given our progress this semester, we will achieve success at this year’s competition and in future years. E. Future Objectives In future semesters, we aim to work on quantitatively testing all aspects of our system, taking advantage of the modularity and sensing capability. We also aim to construct a more efficient boat hull.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 26/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 27
References [1] “PCB Trace Width Calculator”. Advanced Circuits. 2018 https://www.4pcb.com/trace-width-calculator.html
[2] “VESC – Open Source ESC” Benjamin Vedder. 2016 http://vedder.se/2015/01/vesc-open-source-esc/
[3] “PyVESC Documentation” Liam Bindle. 2017 https://pyvesc.readthedocs.io/en/latest/
[4] “Ungava Bay 16” Playak.com. 2010 http://playak.com/buyers-guide/boats-boards/great-canadian-ungava-bay-16
[5] “Metalico Rochester” Metalico Rochester. 2017 http://www.metalicorochester.com/
[6] “IV Curve” PVEducation.org. 2019 https://www.pveducation.org/pvcdrom/solar-cell-operation/iv-curve
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 27/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 28
Appendix A: Battery Documentation A. Material Safety Data Sheets Below is the MSDS for the Genesis batteries used in both Endurance and Sprint.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 28/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 29
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 29/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 30
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 30/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 31
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 31/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 32
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 32/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 33
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 33/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 34
B. Battery Datasheets
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 34/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 35
Note: Our system uses 9 Genesis EP-13 batteries for Sprint and 3 Genesis EP-42s for Endurance, highlighted green in the page below.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 35/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 36
Appendix B: Flotation Calculations
Solar Splash 2018 Rule 7.14.2 Buoyancy of Craft: Sufficient flotation must be provided on board so that the craft cannot sink, even when filled with water. A 20% safety factor must be included in the calculations. Verification calculations must be included in the Technical Report.
Our flotation calculations, as required by the above rule, are shown below. Note: To simplify our calculations, we chose to disregard the volume of all objects on the boat except the boat hull itself. Assuming that all objects are dead weight with zero volume means they provide no buoyancy force, so these calculations have an additional margin of safety.
Endurance Configuration Sprint Configuration Component Weight (lbf) Component Weight (lbf) Boat Hull -197.6 Boat Hull -197.6 Solar Controller 2.8 Solar Controller 0.0 Motor Controller 18.6 Motor Controller 18.6 Telemetry Server 2.4 Telemetry Server 2.4 Telemetry Nodes 10.0 Telemetry Nodes 10.0 Bilge System 5.0 Bilge System 5.0 Solar Panels 41.2 Solar Panels 0.0 Solar Panel Mounts 4.0 Solar Panel Mounts 0.0 Cables 10.0 Cables 10.0 Steering System 10.0 Steering System 10.0 Dashboard 5.0 Dashboard 5.0 Drivetrain 23.3 Drivetrain 23.3 Endurance Motor Assembly 3.4 Sprint Motor Assembly 3.4 Battery Box 35.6 Battery Box 35.6 Endurance Battery Set 100.0 Sprint Battery Set 100.0 Total Force -468.9 Total Force -443.3 +20% Safety Factor -562.68 +20% Safety Factor -531.96 Hull Buoyancy Force 584.87 Hull Buoyancy Force 584.87 Net Buoyancy Force 22.19 lbf Net Buoyancy Force 52.91 lbf
In both endurance and sprint mode, there is a net buoyancy force of 22.19 lbf and 52.91 lbf respectively, meaning that the boat would float even if completely filled with water.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 36/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 37
1) Hull Volume Calculation: Ungava Bay 16 Specifications[4] Weight: 120 lbs Center Line Depth: 16” Center Line Length: 16’ Beam: 48”
To approximate the displacement of the hull, we built a CAD model based on measurements of our hull and the modifications we performed to the hull this year. The hull model is a shell with a thickness of 0.2” (The fiberglass thickness on our boat) and the general hull shape was constructed from the Ungava Bay 16 specifications as well as our own measurements on the hull. For simplicity, we chose not to model the seat, dashboard, or other internal components in our buoyancy calculations.
Hull volume model based on the original boat hull and the buoyancy boxes constructed this year.
Physical Properties for Ungava Bay 16 Model: W eight = 198 lbs V olume = 9 .37295 ft3 Density of W ater = 6 2.4 lbf/ ft3 Displacement F orce of Boat = Density of W ater × V olume of Boat Displacement F orce of Boat = 62.4 lbf/ ft3 × 9.37295 ft3 = 584.87 lbf
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 37/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 38
Appendix C: Proof of Insurance
Solar Splash 2018 Rule 2.9 Insurance : Each participating Team is required to provide proof of general liability insurance from their educational institution or written proof that, as a state institution, they are self-insured. Proof of insurance must be supplied with the Technical Report. Failure to do so will result in a 10 point penalty applied to the Technical Report score.
Our proof of insurance, as required by the above rule, is shown below: (NOTE: This insurance is from the 2018 year. We are working with our school administration to obtain updated insurance certification as soon as possible, and will update Solar Splash HQ when we receive this.)
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 38/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 39
Appendix D: Team Roster A. Undergraduate Roster Bolded names indicate Executive Board members, discussed previously in Project Management. Name Degree Program, Class Year Team Role Chris Dalke Computer Science (2019) President Andrew Gutierrez Mechanical Engineering (2019) Chief Engineer Seth Schaffer Mechanical Engineering (2019) Vice President Martin Barocas Mechanical Engineering (2019) Business Manager, Secretary Nick Davis Psychology (2020) Fundraising Chair Ivan Frantz Physics (2020) Social Chair Leo Orsini Chemical Engineering (2020) Chief Mechanical Engineer Owen Goettler Computer Science (2021) Chief Electrical Engineer Daniel Wong Mechanical Engineering (2020) Member, Mechanical Systems team Heriniaina Mechanical Engineering (2021) Member, Electrical Systems team Rajaoberison Melanie Armenas Mechanical Engineering (2021) Member, Mechanical Systems team Mark Westman Mathematics & Mechanical Member, Hull team Engineering (2020) Daniel Allara Chemical Engineering (2020) Member, Hull team
B. Advisor Roster Name Advisor Role Ethan Burnham-Fay Faculty Advisor Kyle DeManincor Primary Advisor Jill Morris Mechanical Engineering Dept. Advisor Jim Alkins Additional Advisor
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 39/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 40
Appendix E: Telemetry Data A. Description of Telemetry Protocol Each telemetry node is a custom PCB based on ATMega32u4 microcontrollers connected to a variety of peripherals used to collect data. These nodes are connected to the server using USB. Once every second, the server scans its USB ports for new devices. If one is connected, it uses the attributes of the device to choose which communication protocol to use. The server communicates with several types of devices. Most of the devices use a protocol designed by our team (the “Telemetry Protocol”), but the server can also communicate with other devices that have proprietary protocols such as our current sensor. When a node is powered on and plugged into the server, it continuously streams a byte (0x69) requesting to connect and identify as a URSS Telemetry Node device. The server then responds with a confirmation byte and the node will respond with an enumerated DEVICE_ID byte to tell the server how to manage data from it and what data to send back to it. The server then responds with another confirmation byte, confirming that the device ID is valid and then the node is considered “connected” and ready to stream data. If at any point during this handshake, if either the server or node wait for more than 100ms without receiving the data they are expecting, the connection times out and the node will attempt to connect again.
While a node is connected, it will wait for a request packet sent from the server at regular intervals. These request or “heartbeat” packets are sent at 4hz and can also can contain data sent back to the nodes from the server. For example, the throttle data from the throttle board gets packed into binary data on the server and sent back to the motor controller board which will forward that signal to the motor controller. When the node receives this request, it will unpack the transmitted data, pack the current data it has collected from its peripherals and send it back to the server. If at any point during this process the node does not receive a heartbeat from the server or the server does not receive the data it is expecting within a second, the devices will
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 40/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 41
return to the disconnected state and will repeat the handshake to reconnect. This automatic reconnection makes the protocol extremely robust, allowing users to disconnect and reconnect sensors freely without worry of synchronizing the communication protocol as the server and node will automatically handle reconnecting. The data the node sends to the server is broken down into 16 byte packets. While most sensors only send one pack of data per request, some sensors such as the GPS/IMU board have too much data to compress into one packet. These packets follow a common format consisting of a header byte shared between all packets (0xF0), 13 bytes of sensor data, packed according to the device ID, a packet number byte identifying how to unpack the data for multipacket nodes and an 8-bit checksum to ensure that none of the data was corrupted during transmission. Request packets are structured in a similar format.
Telemetry Packet Format 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Header Data Bytes (13) Packet Number Checksum
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 41/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 42
B. Telemetry Node Implementation While the new data transmission protocol implemented by the team is extremely robust and capable of efficiently transmitting large quantities of data at high speeds, it is relatively complex and requires an understanding of hexadecimal, bitmasks, bit shifting and serial communication to implement on the node sensors. This can be a problem for new members in the club who are just beginning to learn embedded programming but still want to help improve the boat. In order to resolve this issue, the team has implemented the telemetry protocol in C++ for Arduino and packaged it as an expandable library that allows developers to easily leverage the capabilities of the protocol with only a few lines of code. The source code for the telemetry nodes can be found here on Github. A parent TelemetryNode object defines the handshake protocol and framework for receiving heartbeat packets and transmitting data. It has a virtual function interface for pack and unpack functions to implement on the child classes for each node. This makes it so that all developers need to do to define new nodes is to inherit from the parent object, add data fields for the values to send and implement a pack and unpack function telling the node how to send and receive data. After the node has been defined, only a few lines of code are required to write an Arduino sketch for the node.
Example implementation of a device node and usage in an Arduino Sketch
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 42/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 43
B. Telemetry Node Data Each telemetry node works with a different set of data. Below is a list of the data sent and received for each node.
Node Name, Description Data Points Received Data Points Sent
Throttle Board None byte throttleIn : boat throttle input used as a throttle input for from a potentiometer value read from the boat an analog pin
Motor Board None uint32_t motorRPM : the RPM of the reads sensors on the sprint sprint motor calculated by using a hall motor and drivetrain. switch uint32_t propRPM : the RPM of the motor scaled by the gearing, computation done of the sensor for easy data processing float motorTemp : the temperature on the motor in ℃ calculated from a thermistor mounted in the sprint motor
Motor Controller Board uint8_t throttleOut: output uint16_t diodeTemp : controller write throttle values to throttle to write diode temp digital potentiometer and uint16_t inVoltage : controller input read sprint motor voltage controller registers uint16_t outCurrent : current through output uint16_t inCurrent : current from batteries, read using outCurrent and controller duty cycle uint8_t dutyCycle : controller dutyCycle uint8_t errorCode : encoder errors from the sprint controller. Each bit is a flag representing and error i.e. under-voltage
Solar Controller Board None float outCurrent1 : current output read current output from from solar charger 1 the solar panels float outCurrent2 : current output from solar charger 2
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 43/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 44
Appendix F: Electrical System Data Below are the schematics for selected telemetry nodes used in our electrical system.
Schematic for the Motor Controller board.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 44/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 45
Schematic for the Solar Charger monitoring board.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 45/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 46
Schematic for the Throttle board. This schematic is different because it was designed using Autodesk EAGLE, before our team transitioned to the use of EasyEDA for PCB development.
Schematic for the GPS/IMU board.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 46/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 47
Appendix G: Power Electronics Data A. Motor Specs Endurance Motor Sprint Motor Model Name Alien Power Systems 6374S Lynch Motor LEM-200 95 Model Notes Sensored BLDC Outrunner Axial-Winded Brushed DC Motor Motor Rated Voltage 42v 48v Rated Torque 9.5 Nm 28 Nm Rated Current 80A 250A Rated Power 2.8kW 10kW Peak Current -- 400A Peak Power -- 18kW Peak Efficiency 95% (Estimation) 92% RPM at Rated Voltage 2520 3888 Speed Constant 60 Rpm/V 81 Rpm/V Torque Constant 0.012 Nm/A (Estimation) 0.113 Nm/A No Load Current 0.65A (Estimation) 6A
B. Motor Controllers Endurance Motor Controller Sprint Motor Controller Model Name VESC 4 Alltrax AXE4844 Model Notes BLDC Sensored/Sensorless Motor DC “Series Wound” Brushed Motor Controller Controller Rated Voltage 8v - 60v 24v - 48v Rated Current 50A 150A Peak Current 240A (5 sec) 400A (2 min) Throttle Input Type USB Serial Variable Resistor (5k Ohm)
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 47/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 48
C. Motor Datasheets
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 48/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 49
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 49/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 50
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 50/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 51
D. Notes on Alltrax Throttle Control Previously, our team used voltage-based throttle control, which had several difficulties. Since our system is 36v nominal, we needed to have an additional DC-DC converter to generate a 5v signal. The grounds on the input and output side of many DC-DC converters are isolated, so when the Alltrax reads the value of the throttle an offset was present based on the difference between the ground references. With the development of our telemetry system, we are using the telemetry network to transmit a digital throttle value. With the new system design, the throttle is sent as a data point into the telemetry system, and is then transmitted to the motor controllers over a serial connection. Since the Alltrax cannot be controlled directly via serial data, we needed to create a circuit which would convert the throttle control signal to one of the inputs of the Alltrax controller. We chose to use a resistive throttle input. In this mode, the Alltrax measures the resistance across a lead and converts this 0-5k value to a throttle percentage. We used a digital potentiometer, which is an integrated circuit containing a resistor array mimicking an analog potentiometer. When a value is written over SPI to the digital potentiometer, it connects/disconnects a series of resistors to produce a given resistance value. The digital potentiometer is built into a PCB with a controlling Arduino. This Arduino continually polls the telemetry network for throttle values, and writes the value over SPI to the digital potentiometer.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 51/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 52
Appendix H: Drivetrain and Steering Data A. Torqeedo Propeller Datasheets
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 52/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 53
B. Drivetrain CAD Drawings
Exploded view of the drivetrain assembly.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 53/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 54
Drivetrain bottom plate, which bolts to the outboard motor unit.
Shaft adapter used to connect the upper unit to the 4-spline outboard motor shaft.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 54/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 55
One of several spacer blocks used to construct the upper unit frame.
One of several spacer blocks used to construct the upper unit frame.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 55/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 56
One of several spacer blocks used to construct the upper unit frame.
A spacer block which mounts under the Sprint motor, adjusting the depth of the shaft to accommodate for the jaw coupling.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 56/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 57
The upper plate used for the endurance motor. The endurance motor is permanently attached to this plate, and the plate is used to switch out motors.
The upper plate used for the sprint motor. Similarly, the sprint motor is permanently attached to this plate, and the plate is used to switch out motors.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 57/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 58
C. Lower Unit Modifications We modified the lower unit by removing the ventilation plate on the lower unit, and filling the void. The ventilation plate prevents air from reducing thrust from the propeller at higher speeds, and we deemed this a reasonable sacrifice in order to allow for larger propellers. In the case that this becomes an issue, we can lower the outboard into the water and/or fabricate a custom ventilation plate which can be mounted higher to accommodate our larger propellers. After this modification, the outboard can now support propellers up to 14”.
Left: The modified outboard lower unit, with the ventilation plate cut back to allow for larger propellers. Right: The outboard motor with frame of the upper unit, mounted on the boat hull.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 58/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 59
Appendix I: Solar Energy System Data A. Solar Charger Datasheets
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 59/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 60
B. Solar Panel Datasheets
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 60/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 61
C. Explanation of Step-Down vs. Boost MPPT Chargers
The power output from a solar cell can be defined by an IV curve showing the relationship between the operating voltage and the output current[6] . Two defining values for a
solar panel are its short circuit current, Isc , and its open circuit voltage, Voc . The short circuit current is the maximum output current of the panel, but exists at 0v, meaning the total power output is zero. Similarly, the open circuit voltage is the maximum voltage difference across the panel, but since the current is zero at this point, the total power output is zero. The best output lies on the curve somewhere between these two points. By altering the load on the solar panel, the position on the IV curve can be changed. The
point where power is maximized is the “Maximum Power Point”, which is some point (Vmp , Imp ) on the IV curve where P=IV yields the maximum power. An MPPT algorithm tracks this point to maximize output from a panel. This is seen in the chart below.
The maximum power point is some some point (Vmp , Imp ) on the curve where P=IV yields the maximum power. Under normal operation, Vbat < Vmp , so the step-down charger works normally.
If the MPPT charger uses a step-down converter and the battery voltage (Vbat ) is greater than the maximum power point voltage (Vmp ), the power is restricted as the converter can only operate at input voltages greater than or equal to the output voltage. This is the case with our
solar panels, as Vmp = ~31v, and V bat = ~42v during charging. This case is seen in the chart below, where the voltage is clamped to the lowest possible value, the battery voltage.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 61/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 62
In this chart, the battery voltage is greater than the maximum power point voltage. Since the step-down converter cannot convert a lower voltage input to a higher output, the voltage is clamped to the lowest possible value, which is the battery voltage.
With a boost converter, the DC/DC converter can operate at voltages up to the battery voltage, which better matches our solar panels and batteries without causing voltage clamping that restricts the output power.
MPPT algorithm with a boost converter, where Vmp < Vbat . The maximum power point is reached, because the boost converter is able to use any solar panel voltage below the battery voltage. This matches our solar configuration.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 62/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 63
Appendix J: Hull Design Data A. Hull Modification Process Below are a selected sequence of pictures from the hull modification process this year.
Original hull with benches, in our shop before modification by our team.
Hull modification work begins. We started by removing the benches. The wood used to build the benches was saturated with water, and there was water pooling on the inside of the sealed compartment.
The hull after benches and transom have been removed.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 63/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 64
To remove any remaining wood and fiberglass, the entire interior of the hull was sanded. This left just the fiberglass shell, which could now be reinforced.
Because we did not have an accurate model of the hull profile, we used templates and trial-and-error to fabricate stringers to exactly the profile of the boat. This involved a large amount of test fitting before the pieces were satisfactory.
Before mounting any of the wooden reinforcements in the boat, they were encapsulated in epoxy, waterproofing them. We also used marine plywood, which has a water-resistant glue.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 64/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 65
We mounted the transom and stringers to the hull using fiberglass strips and epoxy.
Shown are the two main stringers and the transom mounted on the hull.
Next, we mounted the gunwales on the hull. These consisted of two 16’ sections of White Ash. Using a single piece of wood allowed us to smoothly bend it along the hull profile without needing a joint anywhere along the profile which would have complicated assembly.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 65/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 66
Next, we constructed the remainder of the hull reinforcement, made up of the wooden frame and lightweight plywood enclosing the buoyancy boxes.
We painted all the visible surfaces with marine top coat paint. The surfaces inside the buoyancy boxes will be covered with floatation foam, and have been encapsulated with epoxy.
We built a deck covering the buoyancy boxes out of marine HDPE material.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 66/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 67
Next, we put a second coat of paint on the interior, and added a flexible rubber “rub rail” along the gunwale.
At this point, the hull modifications are almost complete. We focused on the interior functionality of the hull, adding mounts for straps which hold down the electronics. Before competition, this work will continue as we apply the finishing touches to the hull and mount the dashboard.
B. Hull Design History Below is a summary of our team’s research into the hull design process. Although we did not build our own hull this year and elected to use a premade hull, the knowledge gained from this process will be of use in future years.
2010 - 2011
The team’s first hull was a premade fiberglass cargo canoe, the Ungava Bay 16, manufactured by Great Canadian Canoes and purchased by the team their first year of existence. This hull is a square-stern canoe, a displacement hull with a weight of ~120 lbs and carrying
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 67/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 68
capacity of ~800 lbs.[7] High stability and effective at low speeds, poor performance in sprint. This hull was modified this year and serves as a good base hull for our team.
2011 - 2014
This design, the first fiberglass boat to be manufactured by the team, was a hybrid planing/displacement hull with a triangular design. Lack of experience in fiberglass construction and a fiberglass-cloth based process meant this hull was extremely heavy and unwieldy, but it served as a valuable experience in learning the techniques of hull construction.
2015 - 2016
The second fiberglass boat to be manufactured by the team, with a “pickle fork” trimaran design. While creative, the short, wide hull caused a large amount of drag. Additionally, the trimaran hull that this design was based of is intended for high-speed racing boats which benefit from the tracking ability, a benefit which did not apply to our team. The hull suffered from cavitation issues due to the wide transom, necessitating a propeller far back from the transom of the hull.
2017
This hull was our team’s first attempt at cedar strip hull construction. This hull was extremely light when unloaded (~20 lbs). The hull had stability issues; the lack of a hard chine and a very round hull profile meant the center of gravity was very high and no significant righting moment was created when the hull was tipped over. Our team was required to include pontoons which caused a large amount of drag. The process of cedar strip hull construction that we learned from the development of this hull is likely to be used in the future due to its simplicity.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 68/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 69
C. Hydrostatics Testing Program As our team experimented with modelling hull designs, we developed a need to quickly determine the hydrostatic properties of the hull models that we developed. Waterline position, stability curves, center of mass, and other properties are valuable during the hull design process as basic metrics for how seaworthy the hull will be. While programs such as ANSYS Fluent are powerful, determining these values through a full fluid simulation is time-consuming and complex. Design software such as Rhino/Orca provides utilities to calculate these values for a hull model, but our team would like to maintain our use of Solidworks for all design work. The team expressed a need for a system which could quickly calculate the following hydrostatic properties for Solidworks-based hull models: ● Total water volume displaced ● Waterline position ● Static Pitch/Roll ● Hull stability curves ● Center of Gravity Based on this need, a team member built a tool used for the calculation of boat buoyancy, stability, and waterline. This tool, called “Float-Util”, can quickly import a model and perform analysis, providing hydrostatic information. Since our team did not end up designing and building a boat hull this year, we did not have a chance to extensively use this tool. However, we can now perform these calculations quickly, which will assist our team in the future.
Screenshots of the Float-Util program in action; pictured right is a boat hull being simulated for roll stability.
1) Workflow: The workflow for using the tool is as follows: 1) A user exports their Solidworks-created hull model to an .STL file 2) In Float-Util, they select the model, enter the desired weight 3) The user can optionally provide a custom COG. If nothing is specified, the COG of the model is automatically calculated.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 69/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 70
Once a model is loaded, the modification of any of the configuration options (weight, COG, etc) will instantly update the output to reflect the new hydrostatic properties. Using this tool, a hull designer can quickly view the consequences of a modification to the hull profile.
2) Mesh Volume & Center of Mass Calculation: The simulation requires calculations of the volume and center of mass for the split hull meshes. We assume that all meshes are enclosed and made entirely of triangle polygons for simplicity. The volume and center of mass calculations both use a similar technique: For each triangle, construct a tetrahedron from the polygon to the origin. The volume and/or center of mass of the tetrahedron can be calculated. Then, based on the direction of the triangle normal (facing towards/away from the origin), the values are added or subtracted to a global accumulator. If the triangle faces away from the origin, this represents the beginning, or "outside", of the solid body, and we add the volume to the accumulator. If the triangle faces towards the origin, this represents the end, or "inside" of the solid body, and we subtract the body spanning from this triangle to the origin. After all the triangles have been iterated through, we are left with values representing the center of mass and volume of the entire body.
3) Iterative Simulation: To calculate the boat’s waterline position, an iterative simulation is run which “floats” the boat in the water using its center of mass, center of buoyancy, and net buoyancy. At each step of the simulation, two operations are performed, which control the position and rotation of the floated body:
● The volume calculation is used to obtain the mass of fluid displaced by the underwater hull segment. The hull weight is subtracted by this value to obtain a delta, which is applied to the boat's Y coordinate. This "floats" the boat vertically, finding an equilibrium for displacement. ● The center of mass calculation is used to obtain the centers for the full body and the underwater body. The center of mass of the underwater buoyancy, also known as the "center of buoyancy", is calculated. A moment is calculated and applied, rotating the hull to find a stable angle. This simulation is iterated until both the position and rotation deltas are significantly small.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 70/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 71
Appendix K: Project Management Data A. Executive Board Roles The executive board is made up of the following roles, elected yearly: ● President: Presides over meetings, maintains organization within the club, and is responsible for guiding the club’s overall mission. ● Vice President: Maintains lab and assists in presidential responsibilities, and keeps up-to-date documentation of policy. ● Chief Engineer: Oversees building of the boat and its different systems, and ensures the engineering teams have the resources needed to succeed. ● Chief Mechanical Engineer: Oversees all aspects of the mechanical systems on the boat. ● Chief Electrical Engineer: Oversees all aspects of the electrical systems on the boat. ● Business Manager: Responsible for managing all finances and transactions, and interacting with club sponsors and the school’s departments. ● Social Chair: Responsible for publicity and maintaining the team’s public-facing website and social media profiles. ● Secretary: Records all topics and meeting minutes discussed during meetings, and sends emails out to members with relevant information. ● Fundraising Chair: In charge of planning, organizing, and running all fundraising efforts.
B. Budgeting Documentation Presented below are the spreadsheets which we used to track expenses for our team. We tracked each purchase using a Google Sheets spreadsheet, and tabulated our total expenses in a cumulative purchase log, as shown below. For each purchase, we used a template form which included details about the items, where the items were purchased, their cost, and our funding source. These spreadsheets were maintained by the business manager.
The cumulative purchase log for the year, showing all transactions and remaining budget.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 71/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 72
A template used for individual purchases. To use this, the team fills out the list and it is added to the cumulative purchase log by the business manager. As the purchase is performed and tracked, the business manager updates the status of this purchase sheet and the master sheet.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 72/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 73
Appendix L: Sponsorship Packet
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 73/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 74
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 74/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 75
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 75/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 76
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 76/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 77
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 77/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 78
Appendix M: Power Budget The power budget is an estimate of how much power we have available to use during endurance mode. The goal is to allow our boat to drive continuously throughout the entire endurance race, with the batteries ending the race drained. To calculate our power budget, we started by finding the effective continuous amperage capacity for our batteries. The 42 Ah rate specified by the battery manufacturer is the 10 hour rate, so the 4 hour discharge capacity will be lower than 42 Ah.
Constant current discharge information for the 42EP battery, provided by the manufacturer.
Based on information available from the manufacturer, we used this data to extrapolate the amperage available continuously from the batteries for 4 H, building an equation shown to the right which provides amperage capacity in terms of discharge time. At 240 minutes (4 hours), the functional discharge capacity of the batteries is 9.623A continuously, or 38.5 Ah. Utilizing this information, we created a spreadsheet calculating our power budget at various sunlight conditions. We assumed an electrical efficiency of 90%, and a range of solar efficiencies from 100% to 0%.
W atts from Battery = Battery V oltage × (Battery Ah Capacity / 4 hours) × Electrical Loss Ratio W atts from Battery = 36 × ( 38.5 / 4) × 0 .9 = 3 11.85 W atts W atts from Solar = Solar Capacity × Solar Efficiency Net W atts = W atts from Battery + W atts from Solar Net W atts with Shore P ower = Net W atts + ((W atts from Solar × 2 hours) × Electrical Loss Ratio)
The resulting spreadsheet is shown below, with a summary of the results:
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 78/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 79
Summary of Power Budget Electrical Efficiency = 90%, Effective Battery Ah = 38.5 Solar Panel Efficiency Target Amps Target Amps w/ 2H Offshore Charging 100% 21.99 28.66 90% 20.66 26.66 80% 19.33 24.66 70% 17.99 22.66 60% (Design Set Point) 16.66 20.66 50% 15.33 18.66 40% 13.99 16.66 30% 12.66 14.66 20% 11.33 12.66 10% 9.99 10.66 0% 8.66 8.66
Based on experimental testing of our solar panels, we chose 60% solar efficiency as our design set point, or approximately 290W solar panel output. We chose this value because we were consistently able to achieve this output during our trials in early spring. The result of this power budget states that we must keep our current draw at 36v at or below 20A, or 720W during endurance.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 79/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 80
Appendix N: Solar Tracker
1) Current Design: Previously, we did not have any solar tracking system for use on land between events. Our team manually propped up the solar panels in the morning and afternoon to point at the sun.
Right: An example from competition of our team propping up the solar panels. We used sawhorses to tilt the panels towards the sun in the morning and late afternoon.
2) Analysis of Design Concepts: Given the importance of harvesting the maximum available solar energy, our team decided that manually adjusting the panels was inadequate. To maximize the amount of power generated, the solar panels must always be perpendicular to the sun’s rays. We designed a solar tracker system that will continuously tilt the panels towards the sun when on land. This project doubled as a skill-building exercise; the project required our members to learn welding skills when fabricating the frame.
Renderings of the solar tracker model, which we plan to use on-shore during competition to improve our solar collection efficiency. The solar tracker is shown with both solar panels attached.
Our solar tracker consists of a motorized pan/tilt mechanism that automatically positions our solar panels to face the sun at all times. To control the angle of the panel, we use a two pairs of photoresistors; each pair controls one of the axes of the panel. The photoresistors pairs are mounted with an angle offset, as shown in the diagram below.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 80/81
5/6/2019 2019 Tech Report - Google Docs University of Rochester, Tech Report ·Page 81
This offset means that the photoresistor measuring a higher brightness value has an angle closer to the sun, and the tracker should tilt towards this photoresistor. To sense these values, the photoresistors are connected to voltage dividers and an Arduino program reads the corresponding voltages. With a high light intensity, the photoresistor will have lower resistance, corresponding higher voltage reading. The opposite occurs with a lower light intensity. The system takes a rolling average of the readings to eliminate noise, and sends a signal to an H-bridge motor controller which drives a pair of linear actuators, tilting the panels. The physical structure of the solar tracker consist of 6061 aluminum square tubing, ASTM A500 structural square tubing and ASTM A36 hot rolled plate. The center column and base of the frame is constructed from steel, and the rotating frame is constructed from aluminum tubing.
3) Design Testing and Evaluation: The solar tracker assembly is currently in fabrication, and we plan to perform testing before and during competition. There are several improvements for subsequent teams to make on the solar tracker to deal with environmental conditions. One concern is wind: When the solar panels are tilted to extreme angles in early morning and late afternoon, they are vulnerable to high winds that may push the system over and cause damage. A safety feature is planned which monitors wind conditions and resets the panels to a neutral angle, minimizing their cross sectional area. Our team purchased a HoldPeak HP-866A anemometer (Shown, right) for this purpose. We chose this device because it outputs wind speed data over a USB serial connection, allowing it to be easily interfaced with the existing telemetry system or a microcontroller in the solar tracker. We have not yet implemented this system.
https://docs.google.com/document/d/1O9zWnMsXp2kttvCnGx9Cloe6BIANx8N5uxl2V2ImnuY/edit# 81/81