# DIS OpenLab reference rover — build and wiring guide

Version 0.1.0-prepilot · 2026-10-15 · `HARDWARE_NOT_BUILT_OR_TESTED`
Hardware-license proposal: CERN OHL variant for future original design files, pending rights review. This document is not a certified electrical or product-safety design.

## Scope and status

This is a physically plausible architecture and validation plan for one shared, low-voltage differential-drive learning station. It is not a purchasing recommendation, verified local price, approved student kit, or completed prototype. No exact sensor, motor, supply, controller module, or breakout vendor has been selected. Price fields remain blank until local quotes are recorded.

The illustrative target of USD 150 applies only to a shared starter station hypothesis. It excludes computers, educator/safeguarding time, venue, internet, freight, duties, safety equipment, tools, failures, and replacement stock.

## Functional architecture

```text
                         USB during setup only
                                 |
                         [controller board]
                    GPIO/PWM |      | sensor input
                             v      v
manual isolation --> [dual H-bridge] <-- [obstacle sensor]
       ^                  |  |
       | motor supply     |  +---- motor B ---- wheel/encoder B
[protected source] -------+------- motor A ---- wheel/encoder A

All logic references and power returns follow the reviewed module schematics.
Motor current never flows through a controller GPIO.
```

The exact ground, regulator, USB, and motor-supply relationship must be reviewed for the selected modules. Never connect two power outputs together merely because they share a nominal voltage.

## Reference driver limits

The TB6612FNG integrated-circuit family is the initial reference class, not proof that an arbitrary marketplace breakout is authentic or suitable. Toshiba's official datasheet lists motor supply `VM` 2.5–13.5 V, logic supply `VCC` 2.7–5.5 V, average output current 1.2 A per channel, and 3.2 A peak under its specified conditions. Peak current is not a continuous design rating. Board copper, connector, thermal, duty-cycle, supply, and motor stall conditions can lower practical capability.

Selection rule: obtain the exact motor's credible stall-current data or measure it under an approved current-limited procedure; choose driver and supply with documented margin; review both channels and thermal conditions; then set lower software command limits during first tests. If this evidence is unavailable, do not energize the motor.

Primary reference: Toshiba, TB6612FNG datasheet: https://toshiba.semicon-storage.com/info/TB6612FNG_datasheet_en_20141001.pdf?did=10660&prodName=TB6612FNG

## Controller variant A — Raspberry Pi Pico / Pico 2

- Exact board variant and headers must be recorded.
- Pico and Pico 2 non-wireless board layouts are broadly pin-compatible, but firmware, processor, ADC behavior, debug connector, and availability are not assumed interchangeable.
- Use an official pinout and select ordinary PWM-capable GPIO after checking every attached function.
- The board is 3.3 V logic. The motor driver logic supply/interface must be compatible.
- Motor power is separate; never source a motor from `3V3(OUT)`, `VBUS`, or GPIO.
- Configure driver standby/enable to remain disabled during boot/reset using a reviewed circuit, not software hope alone.
- A future adapter must implement monotonic timing, bounded PWM, explicit direction/brake truth table, watchdog stop, sensor validity, and boot/reset behavior.

Primary references:

- Raspberry Pi Pico-series documentation: https://www.raspberrypi.com/documentation/microcontrollers/pico-series.html
- Raspberry Pi Pico 2 datasheet: https://datasheets.raspberrypi.com/pico/pico-2-datasheet.pdf

## Controller variant B — ESP32 family

“ESP32” names a family, not one pin-compatible board. Select and record the exact module/development board, SoC, revision, firmware/framework, and schematic.

- Use the selected variant's official datasheet, pinout, hardware guidelines, and board schematic.
- Check strapping/boot pins and reset-time defaults before assigning motor direction, PWM, enable, encoder, or sensor signals.
- Verify 3.3 V logic compatibility and power-source current/transient needs for the selected board.
- Do not copy Pico pin numbers or MicroPython APIs into an ESP32 adapter.
- A future adapter has the same bounded-command/watchdog obligations but a separately reviewed implementation.

Primary reference: Espressif ESP32 Hardware Design Guidelines, GPIO and strapping guidance: https://docs.espressif.com/projects/esp-hardware-design-guidelines/en/latest/esp32/schematic-checklist.html

## Pin-allocation worksheet — must be completed for the exact build

| Function | Pico/Pico 2 GPIO | ESP32 exact-model GPIO | Reset/default state | Verification source |
|---|---|---|---|---|
| Driver standby/enable | | | disabled | |
| Motor A direction 1/2 | | | low/low | |
| Motor A PWM | | | 0% | |
| Motor B direction 1/2 | | | low/low | |
| Motor B PWM | | | 0% | |
| Left encoder A/B | | | input | |
| Right encoder A/B | | | input | |
| Obstacle sensor | | | input | |
| Manual stop feedback (optional) | | | stop asserted | |

Blank cells are a deliberate human/hardware selection gate, not permission to improvise.

## Wiring and assembly sequence

1. Freeze exact part numbers, module schematics, ratings, firmware, and revision in the BOM.
2. Measure chassis mounting space, wheel radius, wheelbase, shaft geometry, and guard clearances.
3. Review driver against motor rated voltage and stall current; review supply/protection against both motors and controller behavior.
4. Prepare an annotated schematic and pin table. A second competent adult reviews it.
5. Mount hardware with power absent. Add strain relief, insulation, guards, and reachable isolation.
6. Verify no short circuits; verify polarity and intended continuity with a meter.
7. Power logic only where architecture permits. Confirm the driver remains disabled through reset and firmware start.
8. With wheels raised or rover physically contained, current limit low, command zero, and adult at isolation, energize motor supply.
9. Test one channel at minimum command, then stop. Check direction, current, heat, noise, and unexpected motion.
10. Repeat for the second channel. Do not proceed on any anomaly.
11. Run watchdog and disabled-state tests before mobile motion.
12. Calibrate only in the approved arena; log every trial and fault.

## Operator power and failsafe checklist

Before power:

- [ ] Adult supervisor and stop operator named.
- [ ] Exact parts/revisions match the approved BOM and schematic.
- [ ] No damaged cell, cable, connector, board, insulation, or loose fastener.
- [ ] Motor stall demand, driver, conductor, connector, and protection ratings reviewed.
- [ ] Correct polarity/continuity independently checked with power off.
- [ ] Rover is guarded/contained; no stairs, roadway, public walkway, or people in path.
- [ ] Wheels raised for first motion; command is zero; driver defaults disabled.
- [ ] Manual motor isolation is reachable and function-tested.
- [ ] Software watchdog and enabled/disabled stop tests have passed on this adapter revision.

During run:

- [ ] One person controls the test; one adult watches the isolation.
- [ ] Start at the approved minimum command and short duration.
- [ ] Stop for unexpected direction/speed, control loss, stall, heat, odor, smoke, noise, loose hardware, or barrier breach.
- [ ] Never reach into moving parts; isolate first.

After run:

- [ ] Set zero/disabled, isolate motor energy, and confirm stopped state.
- [ ] Record configuration, telemetry, duration, current/temperature measurements if approved, faults, and changes.
- [ ] Quarantine suspect parts and mark them unavailable.

## Calibration and validation gates

1. Logic boot/reset keeps outputs disabled.
2. Manual isolation removes motor energy.
3. Watchdog, invalid sensor, and software disable generate zero targets.
4. Left/right direction matches the annotated frame at low command.
5. Encoder sign/count definition is verified if fitted.
6. Wheel radius/wheelbase and command mapping are measured.
7. Three repeated straight and turn trials are logged.
8. Obstacle-sensor raw response is characterized across target surface, angle, ambient conditions, and range.
9. Matched simulation/hardware protocol is reviewed before comparison.
10. High-severity hazards are closed by the accountable technical/safeguarding leads before learners participate.

## Substitution rules

Any controller, driver, motor, wheel, power, sensor, chassis, firmware, or significant wiring substitution creates a new configuration revision. Recheck affected voltage/current limits, boot states, pin allocation, geometry, calibration, stopping behavior, documentation, and procurement evidence. “Compatible” product titles are insufficient.

## Unresolved human/hardware gates

- Select region, exact components, and two comparable supplier quotes where practical.
- Approve the electrical/mechanical design and test procedure.
- Purchase and assemble under local procurement and safety rules.
- Execute empirical current, thermal, stopping, sensor, calibration, repeatability, and fault tests.
- Record reviewer authorization before any student session or public performance/cost claim.
