Preliminary Design Review Report

Transcription

1 Preliminary Design Review Report Submitted To: Inst. Schrock GTA Yang Created By: Team A Nick Bova Bryan Check Tyler Sargent Jordan Scully Brad Sievers Engineering 1182 The Ohio State University Columbus, OH 9 November 2015

2 Executive Summary The first AEV lab performed had each member of the team create their own designs simply based on what each member had thought to be the best approach without having any prior testing or knowledge. Eventually, one design out of the five was picked to be the main design up until this point. This AEV was then analyzed in subsequent labs in order to find the most viable propeller orientation and Arduino commands. In this specific series of labs, one new design (Test AEV 1) was picked, built, and tested on the track against the previous main design (Test AEV 2). Using the techniques developed in Labs 5 and 6 (System Analysis 1 and 2 respectively) and Lab 07 (Design Analysis Tool), both vehicle designs were run on the track with the EEPROM on the Arduino recording data such as power input and distance over the course of the run. The data was then organized and compared using the included MATLAB analysis tool in order to determine which vehicle performed most efficiently. The purpose of developing and testing multiple vehicles is to create an AEV that is best equipped to meet the objectives and requirements that are stated in the Mission Concept Review (MCR). The goal as stated by the MCR is to use these vehicles to aid the rebel alliance in transporting R2D2 units while minimizing operational overhead. The alliance must prepare for war on these remote planets by setting up a supply vehicle that runs along a rail system. The AEVs must be made cost effective and power efficient while also being flexible enough to run on any track and durable enough to handle the extra weight of the cargo while completing its run. In order to ensure that the best possible vehicle is being used for the given task, multiple designs must be tested in controlled environments so that data such as power used and speed can be measured and compared between the different iterations. Since the mission is being completed on a remote planet, energy efficiency is of the utmost importance. From the data compiled, it was determined that the new design performed better than the previous main design. When executing the same code with both vehicles, Test AEV 1 traveled further on the track than Test AEV 2 using the same motor speed while using slightly less energy. This appeared unexpected at first considering the much more simplistic design of AEV 1, however the reduced mass and more even center of gravity ended up contributing to a much more successful run. As a result, it is recommended that Test AEV 1 is used for all future testing. The code that was initially written for Test AEV 2 will be rewritten to better utilize the structure of Test AEV 1 so that it is able to complete the mission at maximum efficiency. It is also recommended that the AEV begin its run in the pusher propeller configuration and switch to the puller configuration when the payload is picked up. The puller configuration was found to be more efficient, so the added weight being towed on the return trip could be offset by this phenomenon. 1

3 Table of Contents Introduction 3 Experimental Methodology Results Discussion..10 Conclusion & Recommendations - Nick Conclusion & Recommendations - Bryan Conclusion & Recommendations - Tyler Conclusion & Recommendations - Jordan Conclusion & Recommendations - Brad Appendix A References

4 Introduction The objective of the lab exercises up until this point have been to design, construct, and test a plethora of Advance Energy Vehicles in order to determine which is most equipped to complete the Mission Concept Review. In Lab 08, two different AEVs were created using knowledge from past labs and tested on the overhead track using an identical Arduino code. The purpose of this procedure was to discover which prototype completed its run more efficiently (used less energy) under the same operating conditions. Energy conservation is an essential aspect of the AEV, for the vehicle will be used to transport R2-D2 units across a planet with extremely limited power availability. Therefore, the quantitative data as well as the analysis skills and understanding of the testing process gained from this activity were quite valuable to the group. The team also learned the importance of using prototype comparison to discern the most effective vehicle, a technique used in the real engineering world frequently. In this Preliminary Design Report, a mid-term progress assessment of the AEV project is provided. First, the experimental methodology of the lab is relayed to the reader including the specific steps followed by the team and the equipment used to complete the tasks. The results are then presented and discussed by the group, followed by the individual conclusions and recommendations drawn by each team member. Finally, an appendix of pertinent figures and tables is present at the end of the report as well as a list of references. 3

5 Experimental Methodology This lab was conducted using a variety materials including two separate AEVs, a computer running Arduino and MATLAB software, a micro USB cable, the overhead testing track, and multiple hand tools. The first step in the procedure was for the team to brainstorm designs for an energy efficient AEV based off of findings from past labs and outside information. After doing so, a member of the group began to assemble the prototype using a small flathead screwdriver and a wrench. As the builder worked on assembling the body, motors, propellers, and wheel count sensors into a functioning vehicle, the remaining four students devised an Arduino code that would move the AEV along the track from its starting point to the first gate. A second portion of code was then developed that commanded the AEV to reverse direction and return to the starting point once again (Figure A1). The reverse(), motorspeed(), brake(), gofor(), and gotoabsoluteposition() functions were all utilized within the script file to accomplish the objective. A clearer picture of the vehicle s path may be seen in Figure 1 below. Figure 1: Test AEV Track Route (Diagram Courtesy of OSU Engineering) Once both the AEV construction and the Arduino code creation were completed, one end of a micro USB cord was plugged into the computer and the other was attached to the port located on the AEV s Arduino controller. The Arduino software containing the code was then opened, the correct COM port was chosen on the computer, and the code was successfully compiled and uploaded to the vehicle (Figure 2). 4

6 Figure 2: Arduino Code Being Uploaded to AEV Following the upload of the code, one student took the AEV to the starting point of the overhead track and positioned the vehicle in the correct orientation, ensuring that the push configuration would be tested in the first half of the run and that the pull configuration would be tested in the latter half. After verifying that a second team member was spotting the AEV at the gate in case of unforeseen issues during testing, the first student flipped the power switch on the Arduino to the ON position and hit the START button. The vehicle completed its run successfully and the team returned to the computer to transfer the data collected by the wheel count sensors into MATLAB, again using the micro USB cable. Using an application developed by the Ohio State Engineering Department, the EEPROM data stored during the run was converted into physical parameters and an efficiency graph was generated in order to aid in analysis (Figure 3). Figure 3: AEV Analysis Tool MATLAB Application All of the above steps were then completed a second time using an identical code, however a different AEV design was constructed and tested in order to determine the effect of geometry and mass on overall efficiency. The results of the two vehicles tests were subsequently compared and discussed by the group. 5

7 Results The goal when designing Test AEV 1 was simplicity. This design consisted of a long rectangle mounted longways with the combination arduino and wheel arm mounted in the center. The motors were mounted in a pushing configuration, one on either end of the rectangle (Figure 4). Test AEV 2 was designed with a small footprint in mind. It had the arduino mounted on the wheel arm, and the body was comprised of two angled wing pieces attached to a small rectangular piece with the motors attached to the underside of the wings giving the vehicle a push orientation when leaving the start (Figure 5). The original five designs the group drafted during the Lab 01 were created with the intent of being simple and light designs. AEV 1 and 2 took these ideas and built upon them with AEV 1 being light and simple to produce and AEV2 being compact and aesthetically pleasing. More detailed depictions of both AEVs may be found in Figures A7-A10 of Appendix A, including orthographic and bill of materials drawings. Figure 4: Test AEV 1 Photo Figure 5: Test AEV 2 Photo The group was expecting design 2 to outperform design 1 as it was smaller and more nimble. While running the experiment the group noted the slow time for Test AEV 2 to reach top speed in comparison to Test AEV 1. When operating in the pull orientation Test AEV 1 was noticeably faster than Test AEV 2 by a large margin. Test AEV 1 also went nearly four meters further while using the same amount of power as AEV2. This observation was surprising to the group, as Test AEV 2 was considered to be the stronger design. This realization changed the groups design process and inclined them to stick to simple solutions to problems, resulting in a decline in the importance of vehicle aesthetics and an increased weight on the efficiency of the designs. Initially the group was concerned with making a design that was unique and looked cool. This design philosophy resulted in a vehicle that exhibited less than stellar performance, and helped the group realize what was truly important. The knowledge gained in the System Analysis Tests 2 was used to upload the data and create graphical representations of the data the group could use to quickly determine the efficiency of each 6

8 design. These gained skills provided a platform on which to develop an opinion on which design fell more in line with the design goals of the team. After the raw data was collected and uploaded to the MATLAB Data Analysis application, a graph of Power (W) vs Distance (m) was generated (Figure 6). Through examination of this graph, it is evident that the earlier claim of AEV 1 traveling further is true. Judging by the geometry of the orange and blue lines, both AEVs were indeed tested using an identical Arduino code, however the second motorspeed() command (second hump of each line) propelled Test AEV 1 quite farther than Test AEV 2. The group determined that this phenomenon was due to the greater amount of power required to move the heavier Test AEV 2 and perhaps that AEV s inferior aerodynamics. Figure 5: Plot of Input Power (W) vs Distance (m) for Both AEV Test Runs After analyzing the graph and gaining all that could be extrapolated from it, the team calculated the amount of energy used by each vehicle in each phase of the run. The AEVs journeys were separated into five sections corresponding to five Arduino commands: reverse(4), motorspeed(4,20), reverse(4), motorspeed(4,20), and brake(4). The distance traveled, time elapsed, and total energy expended were calculated for each phase of each AEV and then all of the information was organized into two tables for the sake of simplicity (Table 3 and Table 4 below). The idea for doing so came from the similar procedure used in System Analysis 2. 7

9 Table 3:Test AEV 1 Energy Phase Breakdown Data Table 4:Test AEV 2 Energy Phase Breakdown Data Thanks to the use of the MATLAB application, no conversion calculations had to be completed by the group to gain any of these values- only simple subtraction was used to find the incremental statistics from the data spreadsheet generated by the software. As a result, no sample calculations or equations are present in this section. Looking at the tables, the first three phases of the run are quite similar for each AEV, the only difference being that Test AEV 1 took approximately.8 seconds longer to reach the same distance of 2.52 m as Test AEV 1 did. This assertion is supported by the graph above, for the plots are coincident for the first three phases. For the purposes of the Mission Concept Review and its objectives, it does not matter how quickly the vehicle completes the objective, only that it does so efficiently. Therefore, both designs were identical in performance for the first three phases. In the fourth phase of Test AEV 1 s run, the vehicle moves 3.56 m while expending just J of energy. Test AEV 2, on the other hand, travels only 1.4 m in this phase and still expends more energy, J. This phase is crucial for examination purposes, for it shows that given the same command, Test AEV 1 was more efficient and traveled a further distance than Test AEV 1 did. As stated earlier, this was not what the group expected to happen. Test AEV 2 had been the design the group had been confident in the past few labs, however it underperformed against its lighter, more simplistic competitor. 8

10 Following the discovery that Test AEV 1 outperformed Test AEV 2, the group next had to decide when to use the push propeller configuration and when to use the pull configuration. Using the information gathered in System Analysis 1, the group knew that the most efficient propeller configuration was the 3 inch blade puller orientation. As one can see in Figure 6 below, this configuration resulted in a high propulsion efficiency for all advanced ratios. As a result, the team decided to use this configuration on the way back to the gate during its final run- the AEV will be carrying a payload at this point and the increased thrust of this orientation will help to offset the mass difference. Therefore, at the start of the run, the AEV will be set up in the pusher configuration. Figure 6: Propulsion Efficiency vs Advance Ratio for 3 Inch Puller Propeller 9

11 Discussion Looking at the concept screening matrix below (Table 3), Test AEV 1 and Test AEV 2 had extremely similar disadvantages and advantages. Two major differences were that Test AEV 1 was seen as more aesthetically pleasing than 2 was and that Test AEV 2 supposedly had superior aerodynamicshowever upon analysis of the results (Figure 5) it has been concluded that the surface area on both front faces of the AEVs is nearly identical and therefore the latter was not a relevant factor. Both AEVs were highly ranked in the Form Factor, Mass and Center of Gravity categories and earned identical net scores in the concept screening matrix. According to the table, both prototypes easily outscored the five design concepts developed in Lab 01 (Figures A2-A6), which was expected. The net scores of both prototypes was 3, while the highest score of any other concept was 2 and the average score of the four original concepts (excluding AEV Concept 4 because it is identical to Test AEV 2) was.5. The team purposely examined what worked and did not work with these AEVs and utilized only the best aspects of each when designing the two for this lab. Table 3: Concept Screening Matrix For 8 AEV Designs Because the screening matrix suggested a virtual tie between the two concepts, a concept scoring matrix was also used to help aid in the team s decision (Table 4). Based off of the table below, the prototypes once again proved to be superior to the concepts from Lab 01 in almost every way. Again, this was expected because the two AEV s used in this lab were designed with the flaws of the earlier concepts in mind. The average score of the early concepts was approximately 2.76, while both AEV prototypes scored above a 3. Test AEV 1 scored higher in the durability and aerodynamics ratings than AEV 2, resulting in a higher overall score of 3.9 compared to

12 Table 4: Concept Scoring Matrix for 8 AEV Designs There were a few opportunities for potential error in this lab. During the testing of the AEVs, the same battery pack was used throughout the entire testing process. Therefore, because the battery was only fully charged for the first test run and was at a different power level after each test, the results from each subsequent test could be slightly inaccurate. This minor inherent error should not have been enough to greatly affect the results, however. Another possible source of error could have come from assembling and reassembling the AEVs. Because the group had to repeatedly switch between the testing of two separate designs, it is possible that the vehicles could have been misassembled at one point. One final source of possible error could have occurred during the AEVs runs. Occasionally, the vehicle snagged on part of the track and had to be pushed along by a team member to continue. This could have thrown off the data, but perhaps could be remedied by increasing the power. Comparing the data from Tables 3 and 4 above, it can be seen that Test AEV 1 used a total of watts of power versus Test AEV 2 using a total of watts. Since the same Arduino code was used to test each concept, Test AEV 1 was slightly more efficient that Test AEV 2. Also, examining Figure 5 above shows that Test AEV 1 travelled a greater distance than Test AEV 2 (about 4 meters) while using the same amount of power. This analysis indicates that AEV 1 is the better option due to the greater energy efficiency and ability to travel further using the same amount of energy, and the concept screening and scoring matrices above back this assertion up as well. Therefore, the objective of this lab was completed and an efficient prototype was found. Using knowledge from System Analysis 1, it was also determined that the AEV should start in the push configuration and return with the payload in the puller configuration. 11

13 Conclusion & Recommendations - Nick This lab has shown that the redesigned AEV 1 outperformed the previous AEV 2 model. As shown in Tables 3 and 4, AEV 1 was able to travel close to three times the distance that AEV 2 traveled while coasting (e.g. when the motors reversed in phase 3) even with both models using roughly the same power. These results show that AEV 1 is more equipped to fulfil the goals that are outlined by the Mission Concept Review, and, as such, the team has decided that it is the best design to use going into the future. Since the team has only enough parts to assemble just one vehicle at a time and, as a result, must disassemble and reassemble the two different designs between runs, there could exist potential error in accidentally assembling the same design differently and, thus, incorrectly between iteration. This possibility is slightly mitigated by keeping a list for all the different parts that each design used, though discrepancies in how each AEV is put together could still exist. This sort of error could affect aspects of the AEV such as mass and center of balance. Considering, however, the relatively simplistic design of both vehicles and the mostly consistent data for each run of each design, the risk of this having happened is low. Still, it is suggested that when doing this type of lab again in the future that the SolidWorks models and sketches be made first so that they may be used as a reference when constructing and the deconstructing the designs. 12

14 Conclusion & Recommendations - Bryan In this lab, the team designed two separate AEV prototypes based off of past concept sketches, system analyses, and graphical interpretation. An Arduino code was simultaneously created that directed the vehicle to move to the first gate and then reverse and return to its starting point. Both AEVs were then tested on the overhead track with this identical code and the data collected during each run was uploaded to MATLAB. After that, the team analyzed the resulting Power vs Distance plot (Figure 5) and Energy Phase Tables (Tables 3 and 4) in order to determine which prototype performed most efficiently. It was discerned that the simplistic, flat, rectangular Test AEV 1 was more efficient than its counterpart, for it completed the same code as Test AEV 2 while expending less energy and traveling a farther distance. It was also determined that the AEV should begin its run in the push configuration and return in the pull configuration. The pull configuration was found to be more efficient (Figure 6), therefore it was concluded that it should be used when the vehicle is carrying a mass in order to offset the added weight. The concept screening and scoring matrices created in this lab supported the team s conclusion that Test AEV 1 was superior as well. As addressed in the Discussion portion of this report, there were three sources of possible error present. The first issue was that the battery lost power every time the AEV was run, so that after the fourth or fifth run the AEV traveled noticeably slower and not as far. To remedy this, a new battery could be obtained by the group after every third run or perhaps the group could only conduct runs when absolutely necessary. The second source of error that could be present was the misassembly of the AEVs. To prevent this error in the future, a second group member could act as a quality control inspector so that the builder has a second set of eyes look over their construction. The final source of error was that the AEV sometimes got snagged at points along the track. To fix this, the AEVs motor power should be increased. The group was able to complete the required tasks with very little issue- in fact the testing was done an entire day early. For future labs, it is recommended that the team moves forward testing Test AEV 1 instead of Test AEV 2. It performed more efficiently and is lighter, which outweighs the unique aesthetic appeal of Test AEV 2. It is also recommended that the AEV begin in push configuration and return in pull configuration, however further testing should be completed using the actual payload to ensure the validity of this suggestion. Overall, the purpose of this lab was met successfully and the students gained valuable experience comparing prototypes and analyzing software-generated charts and data. 13

15 Conclusion & Recommendations - Tyler In the lab the two Test AEV designs were given identical code (Figure A1) and ran to see how much power either design used and how far they could travel on said power. The results of the Test AEV runs determined a very clear winner. The minimalist design of AEV 1 completely outperformed the compact design of AEV 2. Using the roughly the same amount of power ( Watts for AEV 1 & Watts for AEV 2), AEV 1 went significantly further than AEV 2 (Figure 5). The undisputable discrepancy of the two distances made the choice of selecting AEV 1 quite simple. The only potential sources of error in running the AEV trials could have been the beginning placement of the AEV on the track as well as the degradation of the battery life. However, even if there were small differences in the placement of the AEVs at the start of their trials and the battery began to slowly die, there is no way to discredit the overall higher efficiency of AEV 1. Ways to solve these potential errors could be to make a mark on the track and start either AEV from that specific point on both runs. To solve the issue with the battery losing charge over time the, AEV could only be powered on when it is absolutely necessary or the battery could be recharged to a certain level using the smart chargers in the lab. 14

16 Conclusion & Recommendations - Jordan After performing the lab it became immediately clear AEV1 was the better choice for a final design. The results of the lab clearly show that AEV1 travels further while using the same amount of energy as AEV2. This can be attributed to AEV1 being able to coast further during the pause in motors running. This can be seen in Figure 5, as it depicts the usage of energy over distance. The plot of both graphs is uniform for the beginning, but as soon as the brake command is issued AEV1 begins to go a further distance and stays ahead of AEV2 in efficiency. A potential source of error could stem from the Arduino code. This code was rewritten several times in an effort to get it to work correctly, as the group kept miscalculating the number of marks it would take to perform the task. This could have affected the team's AEV run data as different numbers of marks could have been used between the two tests. Another source of error could be the state of charge of the battery as for each design several runs were performed meaning the battery coul dhave drastically different levels of charge possibly affecting the performance. 15

17 Conclusion & Recommendations - Brad Sievers This lab was performed in order to test multiple AEV concept designs and then analyze data collected to determine which design should be used to complete the tasks outlined in the MCR. An Arduino code was written to perform a test run for each AEV concept. Once the designs were tested, graphs of advanced ratio versus propulsion and distance versus power used were created using a Matlab GUI. This allowed for easy data analysis, which indicated that AEV concept 1 was the better option to continue using in the future, due to the greater energy efficiency and ability to travel further while using the same amount of power as AEV concept 2. While some minor inherent error was encountered regarding the charge level of the battery pack, no errors were large enough to greatly impact the data collected. This error could be resolved by charging the battery after a set number of test runs, however this may hinder the amount of testing that could be done, and so is not a likely option. While the data collected during this lab indicates that AEV concept 1 is the better option to proceed with, there are still many other factors to be taken into account that require further testing. For example, this test did not have the AEV pick up the payload. Without more testing, the team cannot determine which design will better complete the task, although the results point towards AEV concept 1. Therefore, it is recommended that AEV concept 1 be the primary design to proceed with, while still testing both designs in upcoming labs. This will allow the team to pick the most efficient vehicle with more supporting data. 16

18 Appendix A Table A1: Group A Project Schedule reverse(4); motorspeed(4,20); gotoabsoluteposition(-430); reverse(4); motorspeed(4,15); gofor(3);. brake(4); // reverse all motors so that the AEV runs in push configuration // run all motors at 20% power until the AEV reaches an absolute position of -430 marks // reverse all motors so that the AEV runs in pull configuration // run all motors at 15% power for 3 seconds // brake all motors Figure A1: AEV Test Run Arduino Code 17

19 Figure A2: AEV Concept Sketch #0 18

20 Figure A3: AEV Concept Sketch #1 19

21 Figure A4: AEV Concept Sketch #2 20

22 Figure A5: AEV Concept Sketch #3 21

23 Figure A6: AEV Concept Sketch #4 22

24 Figure A7: Test AEV 1 Orthographic Views 23

25 Figure A8: Test AEV #1 Bill of Materials Diagram Table A2: Test AEV #1 Bill of Materials 24

26 Figure A9: Test AEV 2 Orthographic Views 25

27 Figure A10: Test AEV #2 Bill of Materials Diagram Table A3: Test AEV #2 Bill of Materials 26

28 References 1. Advance Energy Vehicle Design Project Lab Manual /1182/AEVLab/AEVDocuments/LabManual/AEV_Lab_Manual_Rev_2015_08_07.pdf 2. Technical Communication Guide /resources/techcommguide/tech_comm_guide_rev_2015_07_16.pdf 27