Open Source Solar Car Development at MIT

Last week I had the opportunity to visit the MIT Solar Electric Vehicle Team. Several years ago I worked on a solar car as an undergraduate student, so it was a treat to glimpse the world of solar vehicle racing once again. One thing I have been impressed with about the solar racing community is its camaraderie, which has been an important element since the beginnings of solar racing. Existing teams are excited to see new ones start, and are typically very willing to share some insights into how to be successful in solar racing. The MIT team is planning to take this to a new level of transparency by becoming an open-source solar vehicle team. That is, their documentation and knowledge will be made open to everyone (although the software they use is not necessarily open-source).

While MIT has a very competitive team, education is a top priority for the group and takes precedence over race results. The team is composed of about 25 students, and is based in the MIT Edgerton Center, which is dedicated to hands-on learning experiences for undergraduate students. It’s not just the students designing and building the car who learn something, but sponsors, parents, peers, and the broader community as well. It’s fantastic to see what is possible when bright, creative engineers focus their efforts on energy efficiency. Solar race cars showcase what is possible, and help get us thinking about what we can do to improve current production vehicles. The current MIT solar car can maintain 55 mph on between 600 and 700 Watts of power. That is less than one horsepower! 700 Watts is less than what a hair dryer consumes, and much less than the power of a typical lawnmower. Why the focus on energy efficiency with solar cars? If a solar car is to drive continuously on solar power alone without depleting its batteries, it must use less power to drive than the power it can harvest from the sun. Solar power production is limited by the size of the solar array (which is limited by the size of the car), and the efficiency of the array. Here is a peek at the array on the MIT solar car:

The MIT array is made of monocrystalline silicon solar cells. Common photovoltaic (PV) cells are made from polycrystalline silicon. In these cells you can see lots of little crystals that make up each cell (the photo below is a polycrystalline PV cell). In the array above on the MIT car you can’t see any edges between crystals, because each cell is a single crystal. This makes the cells more efficient. They are also a little bit flexible so they can conform to the curves on the car. The MIT cells are 21% efficient, which is pretty amazing for silicon cells. This means that 21% of the sunlight energy hitting each cell is converted into electrical energy. Satellite-grade PV cells are made from different materials (like gallium arsenide), and can reach efficiencies as high as 40%, but are much more expensive than terrestrial grade silicon PV cells. Click here to learn more about how PV cells work.

The MIT car uses a lithium ion battery pack that is about the size of four regular car batteries. It is made of a large number of laptop batteries wired together with a power management system to keep things under control. Some teams use lithium polymer batteries because of their better power density (i.e., for the same size battery, lithium polymer batteries can output a lot of power), but these batteries have lower energy density than lithium ion batteries. To summarize, if you are comparing a lithium ion battery and a lithum polymer battery that are the same size, lithium ion can hold more energy, but lithium polymer can release its energy faster. The MIT team examined this tradeoff, and learned that for their car and the races they were competing in, lithium ion was the best choice. Tradeoffs like this can be analyzed using the modeling and optimization techniques that I’m addressing in articles throughout this blog. Using quantitative tools like these can help engineers explore design options and make the best choices for their design problems.

In upcoming posts I will describe some of the ways solar car designers squeeze every last bit of energy efficiency out of their cars, and discuss how lessons learned from solar racing can aid advancements in vehicle design for the rest of us.

What do you think about the MIT team’s open development approach, or solar vehicle racing in general?