Team Design Project
Miniature Rover
Team Design Project
A solar-powered rover that had to be good at two very different tests.
This was the main project for my MEMS 0024 class. Each team was handed a rover to design around two tests, a speed run while carrying a set weight and an obstacle course of unknown obstacles, and I led a team of five through it.
The constraint that shaped everything was power. We got a single 10,000 rpm motor with almost no torque, running off a solar panel that peaked around 0.2 watts. We had plenty of speed and nowhere near enough force, so the whole rover became an exercise in trading that spin for usable torque.
Early on we decided to build for stability and the obstacle course rather than chase a fast sprint, betting that a rover solid on both tests would beat one that was great at just one. The other call, which I pushed for, was to get through as many prototypes as we could. None of us had built anything like this, and we were going to learn the most by making things and watching them break.
With that little torque to work with, almost every decision on the rover came back to one thing: the gear train.
The drivetrain was my main area, and it changed more than anything else on the rover. The motor's whole problem was torque, so the fix was gearing: give up the speed we had in excess to buy the force we were missing.
My first gear train was a three-stage, 64:1 reduction packed into a box in the middle of the chassis. The ratio worked on paper, but the box was bulky and got in the way of everything else. So I took a harder look at it, talked through tighter setups with friends who'd built cleaner gear systems, reworked the math, and went back into SolidWorks.
What we ran was a two-stage, 49:1 train mounted to the side of the rover instead of buried in the center. It cleared out the chassis and put noticeably more power at the wheels.
We got three prototypes built in four and a half weeks, and each one taught us what to throw out.
Barely a rover. Just enough of a frame to check that the basic layout held together before committing to it.
Our real attempt at a complete rover. It came out heavy and clunky, built on a solid wooden frame that weighed far more than our tiny motor could ever push.
I handled most of this redesign: the side-mounted gear train, plus a stripped-down version of the chassis cut to the bare minimum we actually needed, roughly 75% lighter than before. After a long day in the shop we had it assembled, ran our own tests, and felt good heading in.
As project lead, most of my time away from the design went to keeping the five of us moving: running the meetings, setting agendas, and checking in often enough to make sure everyone was doing work they were actually into, not just grinding through a task list.
At the competition the rover beat the class average on both tests and came up just short of clearing every obstacle.
I came out of it understanding gear trains and rover design a lot better, but the bigger lesson was the design loop itself: ideate, build, test, tear down, redesign, test again, and how to keep a team of first-timers moving through all of it without losing momentum. Easily one of my favorite projects so far.
