🚗 Twin Wheels

What you're building

Two little motors, side by side, both spinning when you hold the button. Stick a paper wheel (or a bit of foam disk) on each shaft and you've built the drivetrain of a wheeled robot — five levels early.

The whole point of this project is that "one transistor drives one motor" scales cleanly to "one transistor drives two motors". Same brain, doubled mouth. And at Level 9, when you build an actual line-following robot, this is exactly the wiring pattern you'll use — one channel per wheel, one flyback diode per motor, one control line per pair.

The pattern: motors in parallel

Two motors sitting side by side — both hanging off the same collector node, each with its own flyback diode — is the electrical opposite of a series chain. Each motor sees its own independent 4.5 V (minus V_CE(sat)) across its own 30 Ω coil, and each pulls its own current. The transistor's collector adds them up.

Compare against putting the motors in series (nose to tail — motor A's minus lead into motor B's plus lead):

• Series (bad idea for equal motors): each motor sees only half the voltage — about 2.15 V after V_CE(sat). Below the motor's 2 V floor for reliable turning, so they'll stall and buzz. Torque collapses.
• Parallel (this project): each motor sees the full 4.3 V. Each spins at full speed. Combined current: ~286 mA from the AA pack.

Every real robot puts its motors in parallel for exactly this reason. And every real robot gives each motor its own flyback diode, because each coil generates its own independent kick when it stops.

Do the math

Button pressed:

• Base current: (4.5 − 0.7) ÷ 1 kΩ = 3.8 mA.
• Max collector current at that base drive: β × I_base = 200 × 3.8 = 760 mA. We'll need less than half of that.
• Each motor lane: (4.5 − V_CE(sat)) ÷ 30 Ω = (4.5 − 0.2) ÷ 30 = 143 mA per motor.
• Total collector current: 2 × 143 = 286 mA.
• Battery total: ~286 mA motor + ~3.8 mA base = ~290 mA.

Every ceiling to check:

1. 2N2222 collector rating: 600 mA. We're at 286 mA. ✓
2. Base saturation drive: 760 mA capable. We ask for 286 mA collector — plenty of headroom, deeply saturated. ✓
3. Transistor dissipation: V_CE(sat) × I_C = 0.2 × 0.286 = 57 mW. Warm to the touch after a minute; nowhere near hot.
4. AA pack: three fresh alkalines deliver ~500 mA before sagging noticeably. We ask for 290 mA — a hearty draw but well within their comfort zone. Old cells will droop; use fresh.

Button released:

• I_base = 0. Transistor snaps off.
• Both coils' collapsing fields try to push current simultaneously. Each motor's own flyback diode catches its own coil's kick. Coil A recirculates through diode A; coil B recirculates through diode B. Neither one crosses over into the other lane, because each lane is a self-contained loop.
• The transistor's collector rises to ~4.5 V (no path to ground) and stays there, safe.

Why each motor needs its own diode

You might think: two motors in parallel share the collector; can they share one diode too?

Not quite. Here's the subtle bit: the flyback diode's job is to give the coil's own current a place to keep flowing when the switch opens. That current is stored in that specific coil's magnetic field. When the field collapses, the current has to recirculate through that coil — meaning the loop has to include that motor.

With one shared diode across both motors, the diode is still there and forward-biases correctly — but now the two coils are mutually coupled through the shared diode: coil A's collapsing current tries to push through coil B (and vice versa), and the transient is messier. It also works, sort of, in this simple case — but real robot builders always give each motor its own diode. It's cheaper (a 1N4001 costs cents), cleaner, and it's one less thing to think about.

Rule of thumb: every coil (motor, relay, solenoid) gets its own flyback diode across it.

Build it

1. Power rails: battery + to a top row, battery − to a bottom row.

2. Trigger lane: battery + → push button → 1 kΩ → transistor Base.

3. The transistor: flat face toward you, legs are Emitter (left), Base (middle), Collector (right). E → − rail, B ← the 1 kΩ, C into a fresh row A (the "collector node").

4. Motor A lane: motor A between the + rail and row A. Diode A across motor A: cathode (banded end) in the + rail row, anode in row A.

5. Motor B lane: motor B between the + rail (on a different column, but same + rail electrically) and row A. Diode B across motor B: same rule — cathode in the + rail row, anode in row A.

   Both motors, both diodes, share the + rail on the top and the collector row on the bottom. Between those two rails, you have two independent parallel lanes.

6. Press the button. Both motors spin together. Release: both stop together, quietly.

Turning this into an actual wheeled thing

Once both motors spin, tape a small paper disc (or a foam cylinder from a hobby store) to each shaft. Balance the disks so they don't wobble. Then either:

• Stand-alone spinning display: point the shafts up, watch the disks spin. A crude but genuine top.
• Rolling toy: tape or glue the two motors to opposite ends of a cardboard-tube chassis, with the disks acting as wheels and the tube's mouth acting as the axle. Add a paper "tail" at the back so the whole thing doesn't tip over. Press the button — it rolls forward.

This isn't yet a steerable robot — the two motors are wired together, so they always spin at the same speed and same direction. Turning requires driving each motor independently, which is Level 9 territory (with the L293D motor driver chip and two Arduino output pins). But the drivetrain — the "wheels turn when a signal fires them" half — is done, right here at Level 4.

What this opens up

Every wheeled vehicle at Level 9 uses this pattern, twice:

• Right wheel: transistor + motor + flyback diode, controlled by Arduino pin R.
• Left wheel: transistor + motor + flyback diode, controlled by Arduino pin L.

Two of this circuit, side by side, with two independent triggers. That's it. That's the whole physical layer of a robot. Everything else (line sensors, ultrasound, obstacle avoidance) is just deciding when to press R and when to press L.

You already own the physical layer.