Engineering Experience 4: Design a Small Solar Vehicle/Nl/2013: Team AM5
We are Eclipse, a group of students of the International University Group T in Leuven (Belgium). This web page contains all info about our own Small Solar Vehicle (SSV), which we are building for a race at the end of May. You can follow us in our weekly updated blog below. See you at the race...
|Kevin Denis (Team leader)|
|Matthijs Van Dijck|
The first week we started our global work for the project. We handled the management and started thinking about what we wanted to do with our SSV. We chose to build a triangular vehicle made of aluminum with maximal generated sun energy. Therefore we prefer more energy over more aerodynamics. http://commons.wikimedia.org/wiki/File:EE4_Meeting_rapport_1_(2).pdf
This week we calculated the electric properties of the solar panel and thought further about our concept. http://commons.wikimedia.org/wiki/File:EE4_Meeting_rapport_2_(2).pdf
This week we started calculating all the properties of the DC-motor (voltage, power,...) and the transmissing. Secondly we found the mistakes in our Solar panel calculations, the new file can be found below. Moreover we updated the meeting reports so we can use a consequent lay out from now on.
Our temporary report: http://commons.wikimedia.org/wiki/File:EE4_Report_Building_an_SSV.pdf. And the meeting report: http://commons.wikimedia.org/wiki/File:Meeting_rapport_5.pdf
This week we had a meeting with our coach concerning the developments of Case I and Case Simulink. Moreover we had a look at the tasks of Case II. Thirdly we constructed all the parts in K'nex to build them next week. http://commons.wikimedia.org/wiki/File:Meeting_rapport_6.pdf
This week we had to do the test on the theory of the fist two cases. After the test we started working on Case II by building the SSV and starting to work at the excercises (bycicle and collision)this friday we are going to Fablab to make the frame and other components. http://commons.wikimedia.org/wiki/File:Meeting_rapport_7.pdf
This week we tested our SSV on the test track at Group T. The test showed that our SSV wasn't good enough because of the use of K'nex. Therefore we had to deconstruct our SSV and make new plans for the components. We will go to Fablab for new wheels and new axle holders. Moreover we need a new axle and bearing. http://commons.wikimedia.org/wiki/File:Meeting_rapport_8.pdf
Educating: Process report
The history of the SSV
Week 1 - 3
The first real dilemma of the SSV design was the choice for a front wheel driven vehicle or a rear wheel driven vehicle. We consulted our coach what solution would be the best. He explained that a front wheel driven vehicle was much more stable, and had more chances to go straight. Secondly we thought of making an SSV that could be piloted by using the shaft of an old RC-car with its remote controller to control the car. This way it would be easy to steer the SSV when it went too much to the side of the track. When we found out the width of the shaft, we realized it was too small compared to the solar panel, so the idea of steering was abandoned. Subsequently we realized that the orientation of the solar panel was very important. To have a better idea of the best orientation of the solar panel, we decided to send an email to the Royal meteorological institute of Belgium (KMI), to have the angle of the solar rays at the specific date and time of the race. They told us that the angle will be 53°, so that the angle of the solar panel should be approximately 37°.
Week 3 – 8
Our main goal for this project was to invest as little as possible. Our vision of a good engineer was to deliver a maximum output, with a minimal of input. Of course it is easy to make a “super-fast SSV” if you invest loads of money, but we thought that that was kind of a foul solution. We thought about all the materials that we already had at home and that we could use for this project. We decided to use K'NEX for additional structure and for the leading, horizontal wheels and for the shaft. We also decided to use CD’s as wheels because the smaller the ground surface, the less friction. Of course we still had to go to Fablab for the structure of our SSV. There we decided to use MDF (Medium Density Fibreboard) and not Plexiglass. The reason why we prefered MDF over Plexiglass is that Plexiglass is much more expensive, and less handy to work with. We could always drill in MDF, but working with Plexiglass is very clumsy.
In this week we tested our SSV. During the test, we realized that our philosophy of spearing a maximum of costs was a bit too extreme. There was too much friction in the shaft: we need bearings. Our wheels weren’t perfectly fixed to our K’NEX shaft, and weren’t straight enough to make the SSV drive straight forward. Because of this we decided to do the expenses that are needed: we bought bearings on the internet, we bought a shaft in Hubo and made our own wheels in Fablab.
After having fabricated our addition parts, we build our SSV a second time. At the test track we noticed that the inner circles of our wheels (where the drive shaft entered) made in Fablab, weren’t precise enough. The unprecise position arose because the laser first cut out the outer circle and afterwards the inner circle. Because once cut, our wheel moved a bit, the laser cut of the inner circle was not done at the right position. Because of this our SSV couldn’t go straight so it went against the wall very quickly, we even did not arrive at the 10 m mark. We decided to use the “best” wheels that we had at the time: K’NEX wheels. We used the biggest wheels that we had and glued it to the shaft. The result was immediate. With a little aiming, our SSV reached the end of the track! Our SSV was finally finished!
Accessories of the project
When comparing the Gantt Chart with the real progress we notice very few differences. For the most part every person did the tasks he was assigned to in the very start of the project. We do notice differences in the time that was actually spend on various tasks by the different team members. We also implemented some major changes in the broad design of our product. Whilst working on the design we concluded that making a guided vehicle was going to be too hard and time consuming, since we had none of the accessories needed. Therefore we decided to work with a system that would help keeping the vehicle on track when it would collide with the wall. Generally, we expected to have way more time consumed by the actual designing and building then by all the exercises and sub tasks. These extra tasks consumed most of the work because we had to learn how to use new software and methods of calculating various things. We could probably deal more efficiently with some of the problems that we encountered by reorganizing our schedule instead of just trying to catch up. All problems we had now were fixed by working harder and longer than predicted.
In general working as a team went good because every member of the team had a different way of working and all together they support the project really well. Kevin has been the team leader for this project and has done most of the organizing. Thomas and Joris have mainly been contributing on the design whilst Jasper and Matthijs were the main contributors to the big report and the meeting reports. Because the theoretical cases were hard everyone tried to be as much of a help as possible here. The main problem for our cooperation has been the fact that Sven had a lot of work for Industria as fakbar Praesis.
From previous projects this team already had enough experience in making a good work schedule for projects like this. We had a solid WBS and Gantt chart. Joris and Thomas already had a lot of experience working with Solid Works because they learned how to use it in their previous schools. Everyone in the team learned to work with Matlab – Simulink from scratch, all the simulations concerning gear ratio and the usage of the solar panel were learned by everyone. Thomas taught other team members how to work with Solid Works. We also learned how to construct a good Sankey diagram. The most important skills that we did not have and needed for this project we taught ourselves to have a good outcome. Planning is also an important skill that we learned during this project. Planning in the sense that we anticipated busy weeks, and worked more during less busy weeks, so that we could deliver a constant work per week. Dead-line planning is maybe the skill that we insufficiently developed during this project. For Case I and Case Simulink we had to work hard during the last weeks to get to the deadline in time because during the first weeks the progress in learning the needed skills was very slow. Another skill that did cause some problems is that it is not easy to execute enough pressure on a team member (which in our case is also a friend) to get the job done. Sometimes the tasks that were distributed were not finished in time or not done properly.
When we look back to the final product that we made in comparison to what we wanted to make in the plan of approach, we can conclude that we adapted the design only a bit. The design in our plan of approach was rather general, so we were able to keep the key point. Nevertheless we had to change the design in some ways to make the construction of the SSV easier. The first test run showed us the major problems in our design. The wheels and the shaft of the SSV were the main problem in the construction. As told in 'The history of the SSV' we had to change the kind of wheels three times; from CD's to Fablab-wheels to K'NEX-wheels. For the shaft we took a metal shaft in stead of a K'NEX shaft and bearings to make the construction more smooth and stronger.
When we review the process we went through we can decide that we learned a lot. Some new programs were introduced and we got acquainted with some new methods and ideas. Our process was rather fluent and apart from the major changes after the first test run we didn’t have to change our plans very drastically. We worked every week and we were able to finish all deadlines in time.
Advice for future projects
The biggest mistake we made is that we underestimated the difference between a theoretical design and the realistic circumstances. We had to implement a lot of changes to our SSV after the first test run. So we can conclude that we should test more often so that we can adapt the design bit by bit to make it better.
The More circle:
Here are some main aspects for the management of the project.