Rafael Petrosian
Mechanical Engineer
All projects

MotoStudent Race Motorcycle

Rear Suspension Lead · UC Berkeley · Sep 2024 – Dec 2025

I led the 5-person team that designed the back half of a race motorcycle: the swing arm, shock, and linkage that keep the rear wheel planted on the track.

CAD render of the full MotoStudent motorcycle with rider
The full bike in CAD. My team owned everything behind the chassis.

5

engineers on the rear-suspension subteam I led

100s

of linkage iterations before the geometry was final

~97 mm

of designed rear wheel travel

~2×

rise in wheel rate across that travel, concentrated at the end

What I did

  • Led a 5-person subteam designing the rear suspension for the MotoStudent international competition in Spain.
  • Owned the whole rear end: coil-over shock, swing arm, rocker, and linkage, plus how it all mounts to the chassis.
  • Went through hundreds of design iterations in MotoChassis to find a geometry that matched the suspension behavior we anticipated: close to linear through most of the travel, then stiff as a board in the last bit where the forces are highest.
  • Compared solid-body and trellis swing arm designs in Onshape.
  • Worked out packaging, mounting points, and clearances with the chassis team so the rear end fit the frame.
  • Did the full CAD for the rear end and released it to manufacturing.
  • Rebuilt the subteam with committed people after the first group fell through, and started writing everything down.

The hard part

My first subteam wasn't committed, and there was no documentation from earlier years. I carried a lot of the work myself until I rebuilt the team. Now I set expectations and write things down from day one.

How it turned out

The geometry and dimensions I set were carried into the design that went to manufacturing. The full bike wasn't finished before I graduated.

Rear suspension CAD

Close-up of the rear suspension: coil-over, swing arm, rocker, and link
Rear suspension layout used in the final design.
Chassis to swing arm integration render
How the swing arm mounts to the chassis. These dimensions went into the final design.

Engineering detail

The full story for anyone who wants to dig in: design decisions, numbers, and what didn't work.

What the suspension has to do

A race bike's rear suspension has two jobs that pull against each other. Through most of its travel it should be soft and predictable, so the tire keeps contact with the track over bumps and under hard braking and acceleration. At the very end of the travel, where the load is highest, it has to firm up fast so the bike doesn't slam into its limit.

Our anticipated suspension did exactly that: close to linear through most of the stroke, then very stiff in the last bit of travel where the force is largest. My job was to find a linkage geometry that delivered that behavior while still fitting inside the frame.

Hundreds of iterations

In MotoChassis I varied the linkage geometry (pivot locations, rocker and link lengths, shock mounting points), then re-ran the kinematics and checked four plots against our targets: spring and wheel force, wheel rate, motion ratio, and shock displacement.

Nothing in a linkage can be changed on its own. Moving one pivot shifts the motion ratio, which changes the wheel rate (wheel rate scales with the motion ratio squared), which changes how much shock stroke you need, which changes whether the shock still fits the packaging. That is why it took hundreds of passes instead of a handful: each one had to improve the curve without breaking the geometry or hitting the chassis.

The final design was then locked into CAD with the chassis mounts and clearances worked out, and handed to manufacturing.

Final geometry, by the numbers

QuantityFinal design
Rear wheel travelabout 97 mm
Shock stroke over that travelabout 79 mm, nearly linear with wheel motion
Motion ratio (wheel : shock)about 1.31 at full extension, easing to about 1.09 at full compression
Wheel rateabout 50 N/mm at the start, 60 to 65 N/mm through the middle, rising to about 104 N/mm at full travel
Spring forceabout 1,000 N at the start to about 7,750 N at full travel
Force at the wheelabout 800 N at the start to about 7,100 N at full travel

Values read from the MotoChassis plots below.

Kinematics (MotoChassis)

These plots are the result of that iteration. The wheel rate plot is the one I cared about most: it stays gentle through most of the travel, then climbs steeply at the end. That is the stiff-at-the-end behavior we wanted.

Spring and wheel forces vs vertical wheel movement
Spring and wheel forces vs. wheel travel.
Wheel rate vs vertical wheel movement
Wheel rate across travel: gentle through most of the stroke, then a steep rise at the end.
Motion ratio vs vertical wheel movement
Motion ratio, used to size the linkage and choose the shock.
Shock displacement vs vertical wheel movement
Shock displacement vs. travel, confirming the stroke we needed.

Swing arm and packaging

  • Compared solid-body and trellis swing arm designs in Onshape before committing to one.
  • Worked out the mounting points and clearances with the chassis team so the rear end fit the frame.
  • Carried the final geometry into the full rear-end CAD and released it to manufacturing.

What I took from it

The hard problems on this bike were at the handoffs between subsystems: where the swing arm meets the chassis, where the linkage meets the shock. With no documentation from earlier years, I learned to set expectations with the team on day one and write things down from the start.