WELCOME TO THE SECOND annual Ideas for Design issue, celebrating one of our most popular departments. This issue includes additional IFDs, written by readers like you, and feature stories by staff and industry contributors alike. All of these authors got their start as engineering students, learning how to put together their first designs. Today, students still have to complete senior projects as part of their initiation into the real world of engineering—and some of them are winning prizes for their work.
ENGI BOUS PRISE Established last year by Texas Instruments and named after recently retired TI chairman Tom Engibous, the Engibous Prize doles out $150,000 in annual awards to students who design the most innovative electronics systems using analog semiconductors. TI says the prize is the largest of its kind. It’s awarded in three regions of the world—North America, Europe, and Asia.One of the award winners for 2009 is a group of students from the University of Arkansas, who submitted a project called “Photovoltaic Array with Maximum Power Point Tracking.” The system consists of a dc-dc boost converter and H-bridge design with a passive filter and step-up transformer. The dc-dc converter utilizes an MPPT algorithm, charging a dc bus capacitor. The H-bridge maintains a constant voltage on the dc bus capacitor and outputs a pulse-width modulation (PWM) signal, which is then conditioned by a low-pass filter. The output voltage, current, and power are monitored and displayed on an LCD, and a USB drive enables the history of the system to be uploaded to a PC.
Seniors John Damron, Brady Delperdang, Tristan Evans, Jordan Greenlee, and Lauren MeGee all took part in the project. According to McGee, the idea came from American Electric Power via one of the group’s professors. The company is interested in renewable resources and suggested the students develop a system that could take the energy produced by solar panels and transfer that energy to the grid.
When I read their paper, which you can download from the Texas Instruments Web site, I was struck by the variety of talents displayed in regards to both the hardware and software and wondered how the team was assembled. Essentially, McGee said their professor told the class that American Electric Power was interested in sponsoring a project and asked who might be interested in renewable energy and power electronics. As it turned out, these five students were interested in taking on the project, so that was it. The selection process was not based on skill sets, but on interest.
So how did they divvy up the project? McGee said that four of the students had been travelling the electrical engineering path, while one had switched over from computer engineering—good thing, since the project demanded some programming talent. “He had a background in programming that the rest of us didn’t have, so consequently he did the instrumentation part of the project and other software,” she said. “As far as skill sets, none of us really knew what we were getting ourselves into. Only one member of the group had taken a class in power electronics. The rest of us had never seen many power circuits or techniques.”
They did what any good engineer does when faced with a dearth of knowledge. They divided up and began to research their respective parts of the project. McGee and another member studied H-bridge circuits, another took on boost converters, and another filter design. The former computer engineer looked into data acquisition. But the team soon found out that a divide-and-conquer approach wasn’t the best way to attack the design.
“We realized that we all needed to have at least a basic understanding of all the parts,” McGee said. “They weren’t four separate projects. They needed to be integrated as one. That took some teamwork and forced all of us to learn more about how the parts fit together. So we all became familiar with the entire project but had key areas that we focused on.”
Design tools played a big role in the project. The group used Matlab, Simulink, Spice tools, a Web-based heatsink tool, TI’s Code Composer Studio, C, and Visual Basic. As mentioned, part of the project entailed implementing the MPPT algorithm. Maximum power point tracking is a technique that draws the most power possible out of a photovoltaic array. The students used Matlab, Simulink, and Code Composer Studio to program the algorithm on a TI TMS320F2808 DSP, which controlled the power electronics. The students didn’t invent this technique, of course, but wanted to learn more about it by incorporating it into their project.
After the design work was done, the group had to develop a prototype. They selected a printed-circuit board (PCB) manufacturing company, which provided them with a free PCB layout program. McGee said the team populated the boards that were needed to complete the project, using soldering techniques they had to learn as well.
“In the beginning of the project, during the first semester, the hardest part was taking all the information that everyone had, putting it all together and everyone understanding how the system was supposed to work. That took a lot of collaboration and was more research and work for all of us,” McGee said.
“In the second semester, the biggest challenge was testing the project and making modifications. Things always go wrong that you didn’t account for. But I’d also say that’s where we learned the most. Getting in there, learning how to test in stages, making mistakes while testing, and learning from that. It was really difficult, it was a long process, but I would say it’s really invaluable to have. I’m really glad that we all went through that.”
This is a telling comment, as we often learn more when things go wrong than when they go right.