Wiring, Harnesses and Pneumatics

Module 6: Wiring, Harnesses and Pneumatics

Series: FRC Technical Foundations Unit: 3 — Electrical and Power Session length: ~3 hours Prerequisites: Lesson 5

What you should be able to do after this lesson

  • Build a crimp you would trust in an elimination match.
  • Route and label a harness that can be serviced in a pit under time pressure.
  • Isolate an electrical fault with a repeatable diagnostic sequence.
  • Explain how an FRC pneumatic system works and when air beats a motor.
  • Pass an electrical and pneumatic self-inspection.

Wiring is a craft, and it is judged

Lesson 5 was about what the components do. This one is about the part that separates teams: whether the wiring survives contact with a competition.

There is a visible correlation at any regional between the tidiness of a robot’s wiring and how often that robot is on the field. It is not superstition. A robot with a clean, labelled, strain-relieved harness gets diagnosed in two minutes when something fails. A robot with a bird’s nest gets diagnosed in twenty, which is longer than the time between matches.

Rule 6.1 — Wire for the person who will fix it in six minutes with a robot on a cart. That person may well be you, at 4 p.m., between quarter-final matches, with a queue official telling you to hurry.

The craft

Crimps

The majority of electrical failures on FRC robots are bad crimps. Not blown controllers, not dead motors — connections that were never fully made.

A correct crimp:

  • Uses the right die for the terminal. A generic crimper on a ferrule gives a connection that looks fine and is not.
  • Grips the conductor in the conductor barrel and the insulation in the insulation barrel. Both.
  • Passes the pull test. Grip the wire, grip the terminal, pull firmly. If it moves, it was never good. Test every crimp this way, immediately, before it goes on the robot.
  • Has no stray strands outside the barrel, because a stray strand is a short waiting for vibration.

Soldering has its place for small signal connections, but a soldered joint on a wire that flexes becomes brittle exactly where the solder wicks up the strands. For power connections on a moving robot, crimp.

Routing

  • Away from moving parts. Obvious, ignored constantly. Trace the full range of motion of every mechanism, not its resting position.
  • Respect bend radius. A wire bent tightly at the point it leaves a connector concentrates all the flexing at one spot. That spot will break internally, giving you an intermittent fault that is nearly impossible to find.
  • Secure every 100–150 mm. Zip ties or clips. Unsecured wire flails, and flailing wire chafes, and chafed wire shorts.
  • Service loops. Leave enough slack at each device that you can remove it without cutting anything. A harness built to exact length means replacing a motor controller means rebuilding the harness.
  • Separate power and signal where practical. High-current motor leads running parallel to sensor cables for long distances can induce noise. Cross them at right angles rather than running them together.

Strain relief

Every connector needs the mechanical load taken by something other than the connector. A zip tie anchoring the cable to the frame 50 mm before the connector means that when someone snags the cable, the frame takes the force instead of the pins.

Labels

Label both ends of every cable. Use the same names as your CAN ID map and your code constants.

If the code calls it elevatorLeader, the label says elevatorLeader, the CAN map says elevatorLeader. Three different names for the same device is how a pit crew replaces the wrong motor.

Rule 6.2 — One name per device, used identically in the code, the map, and the label.

Diagnostics: a sequence, not a guess

When a mechanism does not move, the temptation is to swap the motor controller. Resist it. Follow a fixed sequence.

  1. Is the robot enabled? Check the Robot Signal Light. This catches more faults than anyone admits.
  2. Does the device appear on the bus? Open the vendor tool and look at the CAN device list. If it is missing, you have a CAN or power problem, not a motor problem.
  3. Does the controller have power? Check the status LED. Every vendor publishes a status-light reference — learn yours.
  4. Is the breaker seated and un-tripped?
  5. Does the controller respond to a direct command from the vendor tool, bypassing your code entirely? If yes, your fault is in software. If no, it is in hardware. This single step splits the problem space in half.
  6. Trace the power path with a multimeter, working from the PD board outward.
  7. Only now consider substitution, and change exactly one thing at a time.

Rule 6.3 — Change one variable per test. Swapping a controller and a cable and reflashing firmware at once means you will never know what was wrong, and it will happen again.

[ANECDOTE SLOT] — An electrical fault that took far longer to find than it should have. Intermittent faults make the best stories here, especially ones where the symptom pointed somewhere else entirely.

Pneumatics

The source FTC curriculum omits pneumatics entirely, because FTC does not permit them. In FRC they are a genuine design option, and they are frequently the right one.

The system

Compressor → storage tanks → pressure regulator → solenoid valves → cylinders

  • The compressor fills the system. It is controlled automatically by the pneumatics module based on a pressure switch. Under the rules it must stop at or below roughly 120 psi, and this is checked at inspection.
  • Storage tanks hold compressed air so you can actuate faster than the compressor can refill. More tanks means more actuations before pressure sags.
  • The regulator drops stored pressure down to working pressure, typically 60 psi. Components at working pressure must be rated for at least 70 psi; components at stored pressure for at least 125 psi.
  • Solenoid valves switch air to the cylinders, controlled from the roboRIO through a Pneumatic Hub or PCM. Single-acting solenoids drive a cylinder one way and let a spring return it; double-acting drive both directions.
  • Cylinders do the work. Force is simply pressure × piston area, so a 1.5″ bore cylinder at 60 psi gives roughly 106 lbf of push.

Safety: a relief valve and a manual vent valve are required, and the vent must be operable without tools. A pressurised system is dangerous even when the robot is off. Nobody works on pneumatics until the system has been vented and a gauge reads zero.

When air beats a motor

Pneumatics are binary, fast, and hold position without power. That combination is genuinely hard to beat for:

  • Deploying and retracting an intake.
  • A gripper that is either open or closed.
  • Shifting a two-speed gearbox.
  • Anything that must hold a position all match without drawing current.

Pneumatics are the wrong choice when you need positional control — variable travel, controlled speed, feedback. A cylinder goes to one end or the other. If you need to stop halfway, use a motor.

The other cost is a fixed budget: compressor weight, tank weight, and plumbing, for a system that adds maybe 20 N·m of useful work. If you are only using one cylinder, that overhead is usually not worth it.

Rule 6.4 — Pneumatics for binary, motors for positional.

🔧 Exercise 6.1 — Competition-Grade Harness Build

Time: 120 minutes. Equipment: A chassis or wiring board, wire, terminals, correct crimp tools, labels, zip ties, a multimeter.

Working in pairs, wire a complete drivetrain from battery to four motors: main breaker, PD board, four channel breakers, four motor controllers, CAN chain, roboRIO power, radio.

Requirements — this is scored, not just completed:

  • Every crimp passes a pull test, tested in front of a peer.
  • Every cable labelled at both ends.
  • Every connector strain-relieved.
  • Service loops at every device.
  • Continuity and polarity verified with a multimeter before the battery ever goes in.
  • Frame isolation verified: greater than 120 Ω between either battery post and the chassis.

Then have a different pair inspect and score the harness against the checklist. Peer inspection is the point.

Evidence of learning: A scored inspection sheet, a wiring diagram matching the physical build, and a photograph of the finished routing.

🔧 Exercise 6.2 — Electrical Debugging Escape Room

Time: 60 minutes. Equipment: A working robot, prepared faults, a stopwatch.

Adapted directly from the source curriculum, and one of the best exercises in it.

A mentor injects a labelled fault while students are out of the room. Good faults:

  • A CAN wire disconnected mid-chain.
  • Two devices given the same CAN ID.
  • A motor controller’s power leads reversed.
  • A partially trip-seated breaker.
  • An encoder cable unplugged.
  • A deliberately bad crimp that passes visual inspection and fails a pull test.
  • A depleted battery.

Teams have a strict time limit and must narrate their diagnostic sequence out loud as they work. The narration is what is being assessed, not the time.

Evidence of learning: A written fault tree produced afterwards, mapping symptom to the checks that isolate it. Keep this document — it becomes your competition troubleshooting guide.

🔧 Exercise 6.3 — Pneumatic Circuit Build

Time: 60 minutes. Equipment: Compressor, tanks, regulator, pressure gauges, a solenoid, a double-acting cylinder, tubing, safety equipment.

Build a complete legal circuit on a bench board. Then:

  1. Verify the compressor cuts out at the correct pressure.
  2. Verify the regulator holds working pressure.
  3. Demonstrate the manual vent bleeding the system to zero.
  4. Measure how many actuations you get from a full system with the compressor off — this is the number that matters in a match.
  5. Calculate the theoretical cylinder force and measure the actual force with a spring scale.

Evidence of learning: A plumbing diagram, an actuation count, and a force comparison between calculation and measurement.

Further reading

Next Lesson

The hardware is now capable of moving. Lesson 7 — WPILib and Software Architecture is about writing code that will still make sense in week five.