Electronics · Semester 1 — Analog · Lesson 7 · 55 min · build 2026.09.23-1644
The big idea
The transistor lets a small current control a big one. Every automatic thing electricity does — sensing, deciding, computing — starts with the transistor family. Ours is the classic; your laptop packs billions of its cousins.
Lesson 6, no peeking: charge a 1000 µF capacitor through a 220 Ω, meter-prove ≈ 5 V, disconnect the supply path, and light an LED from the stored charge. Stripe to ground — say it out loud. Done when you've seen the fade.
Three legs, and each has a name and a job. The big current flows in at the collector and out at the emitter — but only when a small current is flowing into the base. Think of a faucet lever once: the effort at the handle is not the water. From here on, the real words: base current permits collector current — up to the transistor's gain times bigger. The collector's own circuit decides how much of that permission actually gets used.
The base is a diode
Inside, the base-to-emitter path behaves like a small LED: it starts conducting near ≈ 0.7 V and holds close to that as the current climbs. So the naked-LED rule applies here too: the base never connects straight to 5 V — it always gets its own resistor. A bare base across the supply is a huge current and a cooked transistor.
Symbol vs. part: match by name, not by shape
The drawing and the part will never look alike — schematics always point the collector toward + and the emitter toward ground, because the whole page reads downhill, while the package reads E-B-C in a flat row. Don't make the build imitate the picture. Find each wire's letter in the schematic and take it to that leg's row on the board.
On, off — and the dangling wire? Probably off too. But look closely at what "dangling" means: nothing is holding the base off. Touching + is a command; touching ground is a command; dangling is no command at all, and an uncommanded input is at the mercy of static, damp air, and luck. It behaved today; never design a circuit that needs it to. The rule, from here through the whole digital semester: an input you mean to be off gets wired to ground — told, not left to hope.
Still: you just operated a switch with no moving parts — flipped by a wire carrying a small fraction of the current it commands. Where lesson 2's slide switch needed a thumb, this one needs only a whisper of current — and anything that can make a whisper can now be a thumb.
Black probe parked at ground. Red probe on the transistor's own legs — predict each before you measure:
| Measurement | Control at +5 V (LED on) | Control at ground (LED off) |
|---|---|---|
| Collector leg to ground — across the "switch" | ||
| Base leg to ground (the transistor side of the 4.7 kΩ) |
What the numbers say
On: the collector reads ≈ 0.2 V — the transistor is almost a closed switch, almost a plain wire to ground. Off: it reads ≈ 5 V — an open switch, and the whole supply piles up across the open gap, exactly like the dark circuit in lesson 4. The base, when on, sits at ≈ 0.7 V: the base's own voltage drop — same idea as an LED's, smaller.
How much bigger is the permitted current? The meter can ask: the dial position marked hFE and the little socket with lettered holes. Watch the front table: teachers will power off, borrow two transistors from the room, and socket them — same part number, same bin, and the numbers land anywhere from about 100 to 300. That's the gain — engineers write it β, "beta" — and each reading is one sample, taken at the meter's own tiny test current, not a constant of nature.
Design for the weakest
If β is a lottery, how does anyone design with it? Plan on the smallest number the bin might hold. Our switch feeds the base several times the minimum whisper a weak part would need — so every transistor in the bin slams fully on, and a strong β is a bonus, never a requirement. (Want your own part's number? Stretch #1.)
In lesson 5, the pot dimmed the LED smoothly, end to end. Feed the pot's tap to the base instead and something different happens:
Most of the turn does nothing — dark, dark, dark — then the whole performance arrives in one narrow slice: a faint glow, a quick climb, full brightness. After that, more turning changes nothing again. Below the base's ≈ 0.7 V door, no whisper gets in. Inside the slice the transistor really is dimming — where the slice sits depends on your pot and your part. Past it, saturation: fully on, permission to spare. Where lesson 5's dimmer spread its change across the whole knob, this circuit crushes it into a sliver: almost off-or-on. A circuit that squeezes a smooth quantity toward yes-or-no is making a threshold decision, and that word is next lesson's whole story.
Lesson 6's capacitor could light an LED for about a second. Watch what the transistor does for it:
The LED holds on — for on the order of ten seconds — then dies. Same capacitor, same stored energy as lesson 6's one-second fade. The difference: the LED's light is paid for by the supply now, through the collector; the capacitor's store goes only to the base's whisper. A small store, trickled out slowly, commanding a big bright current: that is amplification, and you just watched it.
The timer schematic one more time — but now it reads C = ▢, base resistor = ▢. Pick them so the LED stays on as close to five seconds as you can get after the charge wire lifts. Predict which direction each lever pushes before you swap anything. Closest table wins.
Go further — for the fast and the curious
1. Sample your own β: power off, pull your transistor, seat it in the meter's NPN socket holes E-B-C, read the number, compare across the table. Then reseat it — flat face toward you, E-B-C, said out loud. 2. Swap the red LED for blue in the collector loop. Predict the collector-to-ground reading when on, and the brightness, before you look. 3. Replace the 4.7 kΩ base resistor with 1 kΩ, then 10 kΩ — what changes, in the pot build and in the timer? 4. Bold move: make yourself the base resistor — control wire in one hand, a jumper from the + rail pinched in the other. Your ≈ 100 kΩ body is a legal base resistor; the glow may be faint. (A Darlington pair — two transistors stacked so their gains multiply — makes this trick effortless; ask us about it.)
The math, for the curious — always optional
Your switch by the numbers. Fully on, the loop drops 2.0 V (LED) + 0.2 V (transistor), leaving 2.8 V across the 220 Ω:
The minimum permission, at a weak β of 100:
What the 4.7 kΩ actually delivers:
— about seven times that minimum, on purpose. The surplus is why the whole bin works: the switch saturates whether your β is 100 or 300. The Transistor's Arithmetic sheet on the math shelf runs the whole design, including how the 4.7 kΩ was chosen.
On your card: 1. draw the transistor switch from memory — symbol, all three legs labeled, base resistor in place. 2. One sentence: what does the base current actually do? Hand it in at the door.
Then lesson 6's teardown ritual: bare 220 Ω across the 1000 µF until the meter calls it empty. Retire the drained caps and the pot to the box — but leave the switch standing: LED, 220 Ω, transistor, base resistor, control wire free. Next lesson hires it exactly as it stands.