Formula Student: 97 Iterations to a Race-Car Chassis

Adapted from my college blog, Life of a Mechanical Engineer.

Formula Student is a competition where university teams design, build and race a single-seat race car. Ours was the college’s first attempt. I was the head designer on a 20-person team, responsible for the chassis and for making sure every department’s design could actually be manufactured.

The problem: three requirements pulling in different directions

The chassis had to satisfy three things at once:

• Compliance with the Formula Student rulebook, including its roll-cage specifications,
• minimum weight, because every kilogram costs performance, and
• structural integrity, because it protects the driver.

Make it lighter and it gets weaker. Make it stronger and it gets heavier. Every change had to be checked against the rules again. I read the rulebook at least three or four times, and redesigned the chassis from scratch at least ten times, before a design held up.

How we got there

The project ran for more than eight months. Design alone took five months and nearly 97 iterations, modelled in Fusion 360 and tested through hundreds of hours of simulation.

Thousands of hours of designing and hundreds of hours of simulation gave us our final design.

It wasn’t the optimal chassis, but for a first-time team it was a pretty decent one.

What it taught me, looking back

At the time I thought of it as a design project. Today I recognise most of it as program management:

• Constraints first. The rulebook was a non-negotiable requirement, much like a regulatory standard in product development. Designing without re-reading it was the fastest way to waste an iteration.
• Iteration is the plan. Nobody could have specified the final chassis up front. Ninety-seven versions was how the team learnt what “good” looked like.
• Manufacturability is a cross-team handshake. A design is only finished when the people building it can build it. Coordinating that across departments was as much of the job as the CAD.
• Twenty people, one deliverable. Keeping a large volunteer team aligned for eight months was my first lesson in delivery, long before I had the vocabulary for it.