The core trade-off
An FRC motor like a NEO spins near 5,676 rpm with no load but produces very little usable torque at that speed. A drivetrain wheel needs to turn at maybe a few hundred rpm with a lot of torque to move a 100+ lb robot. A gear reduction converts the motor's high-speed, low-torque output into low-speed, high-torque output.
Conservation of power: ignoring friction, output power equals input power. So if you reduce speed by a factor of N (the gear ratio), you multiply torque by roughly N. Speed down, torque up, by the same factor.
Gear ratio basics
A gear ratio is driven teeth ÷ driving teeth (output over input). Example: a 12-tooth pinion driving a 60-tooth gear is 60/12 = 5:1. The output turns 5x slower and roughly 5x stronger.
You often need more reduction than one gear pair can give, so you stack stages: a two-stage gearbox of 4:1 then 3:1 yields 4 × 3 = 12:1 overall. Multiply the stages.
Picking a drivetrain ratio
Drivetrain reductions are chosen to hit a target free speed (theoretical top speed) of roughly 12–18 ft/s depending on the game, while leaving enough torque that the robot can accelerate and push without stalling. You can compute this with online calculators (e.g., ReCalc at reca.lc, or JVN's mechanical design calculator) by entering motor, gear ratio, wheel diameter, and robot weight; ReCalc also flags whether your chosen current limit keeps you out of the brownout/pushing-stall danger zone.
COTS gearboxes
You rarely cut your own gears. Common COTS options:
- VEXpro / WCP VersaPlanetary — a modular planetary gearbox with interchangeable stages; the stages combine into a very large number of ratios, and it bolts to many FRC motors (BAG, Mini CIM, RS-550, 775pro, AM-9015, CIM). It mounts on a 2" bolt circle, the same as a CIM, so it fits anywhere a CIM does; NEO/brushless mounting uses a vendor or third-party adapter plate.
- AndyMark Sport, Flyer, and ToughBox gearboxes — drivetrain and mechanism gearboxes; the Flyer is designed to run as an overdrive or underdrive.
- REV MAXPlanetary and WCP gearboxes integrated with their structure systems.
Don't over-reduce
Too much reduction makes the robot painfully slow; too little makes it fast but unable to push and prone to stalling (which spikes current and risks brownouts). The right ratio is a balance you tune per game and per robot weight.
Key takeaways
- Gear ratio = driven teeth ÷ driving teeth; reducing speed by N multiplies torque by ~N
- Multiply stages: a 4:1 then 3:1 gearbox is 12:1 overall
- Use COTS gearboxes like the VersaPlanetary or AndyMark ToughBox/Sport rather than cutting custom gears, and pick a ratio that balances speed against pushing torque
Lesson quiz
RequiredAnswer all 3 questions correctly to complete this lesson.
01.An FRC motor spins very fast but with relatively low torque. Why do mechanisms almost always add a gear reduction between the motor and the output?
02.A gearbox has a 10:1 reduction. If the motor spins at 6000 RPM, what is the approximate output speed?
03.What happens to a mechanism's output torque when you increase the amount of reduction (ignoring efficiency losses)?
Answer every question to submit.
All 47 lessons in Mechanical, Build & Pneumatics
- Not started:Mini-Project 1: A Single-Jointed Arm From Math to Motion
- Not started:Mini-Project 2: A Two-Stage Cascade Elevator
- Not started:Mini-Project 3: A Velocity-Controlled Flywheel Shooter
- Not started:Mini-Project 4: A Pivoting Roller Intake
- Not started:Mini-Project 5: Integrating a COTS Swerve Module
- Not started:Pneumatics Won't Fire: A Full Diagnostic Tree
- Not started:The Robot Won't Drive Straight (and Other Drivetrain Sins)
- Not started:Gearboxes That Grenade and Fasteners That Vibrate Loose
- Not started:Closed-Loop Mechanisms That Oscillate, Sag, or Stall
- Not started:Field-Ready Reliability: Inspection, Spares, and the Pit Checklist
- Not started:Characterizing Any Mechanism with SysId
- Not started:Simulation-Driven Design with WPILib Physics Models
- Not started:Motion Profiling and Superstructure Coordination
- Not started:Designing for Weight, Stiffness, and Manufacturability
- Not started:Case Studies: Learning From Open Alliance Robots