SAE Supermileage and Shell Eco-Marathon
Building A Super Mileage Vehicle with the Cal Super Mileage Team SMVBy Clarissa Harrison
Every year we attempt to do what hundreds of professional teams competing for the X-prize failed to do, build a Super Mileage Vehicle that meets the technical requirements, passes safety inspection and makes it through the required course.
We are the student engineering competition team from UC Berkeley. The team's peak was in 2002-2004, when we consistently ranked in the top four at the Society of Automotive Engineer's annual design competition. Since then bad luck and Murphy's law have haunted our footsteps and kept us out of contention. This year, tapping designs from the glory days, new thinking from lessons learned, and a few Jameco components, we have a worthy competitor.
One of the major changes we are making this year is to bring back the heavily-modified engine based on a 150cc flathead Briggs and Stratton crankcase. Not only will this alleviate the power problems we encountered in last year's race, it also makes us eligible for two competitions, the Shell Eco-marathon and SAE Supermileage.
In addition to the mechanical challenges presented by refurbishing an engine, we will be facing the conversion of our home-built electronic fuel injection system from the 25cc Honda it was designed for to the larger capacity 50cc engine. This should be as simple as generating a new map, however there are the issues of re-wiring sensors, sourcing a new injector, and dealing with scalability issues.
SMV's fuel injection system runs code in C on a LM3S8962 Luminary Micro evaluation board. Using sensor data, it calculates the optimum fuel volume based on known engine characteristics and adjusts injector open time accordingly. The supplementary system to this is the dynamometer, which runs the engine through a range of loads and throttle variances in order to measure power, and fills out a volumetric efficiency map based on RPM and manifold pressure.
In more specific terms, we measure manifold absolute pressure (MAP) via a pressure sensor on the intake manifold, the volume of the combustion chamber is known, and a temperature sensor on the circuit board is an approximation of the temperature of intake air. A Hall effect sensor on the flywheel of the engine gives RPM, and the unknown that has to be determined experimentally is volumetric efficiency, the ratio of pressure inside the cylinder to pressure in the manifold. Once that map has been generated by the dyno, the ideal gas law, PV = nRT, is used to solve for volume of air being taken in at each cycle. A balanced chemical equation for the combustion reaction gives the correct ratio of fuel to air. Using flow rate of the injector the correct open time is calculated, divided by two to account for two injections per combustion cycle, and that width pulse sent to the injector.
A little technical, but knowing the logic behind the code is the easiest way to understand the importance of every parameter. The dyno also uses a narrow band oxygen sensor to self-adjust injection times to solve for volumetric efficiency. One of the longer-term goals is to implement a wide band oxygen sensor, and find one that updates fast enough to be used in conjunction with the microcontroller board when the engine is running in the car.
You can follow the team at smv.berkeley.edu. They faced their first challenge on April 14th in Houston Texas. Here's a brief update on their first event this year:
The competition did not go as planned. On arrival we were fairly confident in our work and expected to get past tech inspection by Friday. The engine we planned to run was not ready in time and so we used a stock Briggs & Stratton. The Briggs & Stratton generates 3.5 HP which is a bit much for our drive train which caused what seemed like a continuous list of failure issues. In the end we attempted 5 runs and completed none of them. On our last run we were seriously confident, but a Canadian team cut us off in a corner and in an effort to avoid a collision we ended up rolling our car. It's impossible to say what would have happened if that run continued, but the car as it stands now is sound, and testing this weekend will see whether that would have been our completed run.
Clarissa Harrison is a senior studying mechanical engineering at UC Berkeley and the team lead of the Super Mileage Vehicle Team.
If you have an electronics story you'd like to share, please send it to [email protected].