The LED: a diode you can see
A light-emitting diode is, first and foremost, a diode. Forward-bias it and current floods across the junction exactly as you saw in guide 1; the new trick is that as electrons fall across the gap they hand off their energy as photons, so the LED glows. Reverse-bias it and, like any diode, it blocks — and stays dark. So everything you already know carries over: it has an anode and a cathode, a knee, and a forward voltage drop you must pay to make it conduct.
The one big difference is the size of that toll. A plain silicon diode drops about 0.7 V, but an LED's drop depends on its colour — roughly 1.8 V for red, 2.1 V for yellow, and 3.0 to 3.4 V for blue, green, or white, because bluer light means a higher-energy photon and a wider band gap. Whatever the colour, the I-V curve is still that brutally steep exponential: once you reach the knee, a tiny extra voltage sends the current — and the brightness, and the heat — soaring.
Sizing the current-limiting resistor
That partner is the current-limiting resistor. Put a resistor in series with the LED and the two share the supply: the LED stubbornly holds its forward drop, and the resistor swallows whatever is left over. Because a resistor obeys Ohm's law — a clean straight line, not an exponential — it sets the current gently and predictably, exactly the design lesson from the diode-model worked example. The LED's own drop barely matters; the resistor is in charge.
- Pick your target current from the LED's datasheet — say 10 mA for a comfortable, efficient brightness on an indicator LED (many run happily at 2 to 20 mA).
- Work out the voltage the resistor must drop: supply minus LED drop. For a 5 V supply and a red LED at about 2 V, that is 5 - 2 = 3 V.
- Divide by the target current: R = (5 - 2) / 0.01 = 300 ohm. The nearest standard value is 330 ohm, which gives (5 - 2) / 330 ≈ 9 mA — close enough, and erring slightly low is safe.
- Check the resistor's power: P = (5 - 2) times 0.01 = 0.03 W = 30 mW. A common 1/4 W (250 mW) resistor laughs at that, so it runs cool. Done.
Two honest cautions. First, raise the supply and the resistor's power dissipation climbs fast — drive that same 10 mA LED from 12 V and the resistor must drop 10 V and burn (12 - 2) times 0.01 = 0.1 W, so check the rating every time. Second, never gang several LEDs in parallel behind a single resistor: their drops differ slightly, the lowest-drop one hogs the current and the rest stay dim. Give each LED its own resistor, or wire them in series so they share one current by definition.
The flyback diode: catching inductive kickback
Now a protection job. Recall that an inductor is a flywheel for current: it resists any sudden change in the current through it. A relay coil, a solenoid, a motor — each is a big inductor. While current flows it stores energy in its magnetic field, and the flywheel is spinning. The trouble comes when you switch it off.
An inductor's voltage is V = L times (rate of change of current). Open a switch and you try to force the current to zero almost instantly — a colossal rate of change — so the coil answers with a colossal voltage spike of the opposite polarity, often hundreds of volts, as it strains to keep its current going. That spike is the inductive kickback, and it will gladly punch a hole through whatever transistor or switch just opened, or arc across mechanical contacts and pit them. The fix is to give that flywheel current a gentle place to coast to a stop.
That place is the flyback diode (also called a freewheeling diode), wired straight across the coil but reversed — cathode to the positive end. While the switch is on, the diode is reverse-biased and invisible. The instant the switch opens and the coil tries to fling its end positive-the-other-way, the diode becomes forward-biased and offers a loop: the flyback diode hands the current a path to circulate harmlessly through the coil's own resistance, decaying smoothly instead of spiking. The kickback is clamped to about one diode drop above the supply — the difference between a 0.7 V nudge and a 300 V execution.
Clippers and clamps: reshaping a wave
The same one-way valve can reshape a signal on the fly. A clipper (or limiter) puts a diode in parallel with the signal, aimed at a reference level. As long as the signal stays below that level the diode is reverse-biased and does nothing; the moment the signal tries to climb past the level plus one diode drop, the diode conducts and shunts the excess away, flattening the peak. Point two diodes at the two supply rails and any input that overshoots is clipped to roughly a diode-drop beyond each rail — a tiny, instant clipper that guards a delicate input, the very same idea as the ESD-protection clamp diodes built into nearly every chip pin.
CLIPPER CLAMP Job chops / limits the peaks shifts the whole DC level Topology diode in PARALLEL, to a rail series CAP, then diode to a rail Output flattened where it clips same SHAPE, moved up or down Reads as "go no higher than here" "sit your low peak right here" Typical use input protection, wave shaping DC restorer, level shifting
A clamp does something different: it shifts the entire waveform up or down without changing its shape. Put a capacitor in series with the signal, then a diode from the output node to a reference. On the first peak that forward-biases the diode, the capacitor charges up to the peak value and then holds it, like a memory of how far the wave swung; from then on it adds that stored offset to every later cycle. The classic result is a clamp (or DC restorer) that pins the low peak of an AC wave to near 0 V so the whole signal rides above ground — exactly how old video circuits re-established the black level. Clip to limit a swing; clamp to relocate it.
Schottky, protection, and closing the rung
One cousin keeps showing up because it fixes the diode's two everyday weaknesses. The Schottky diode trades the PN junction for a metal-to-silicon contact, giving a low forward drop of about 0.2 to 0.4 V and very fast switching with almost no recovery delay. That makes it the default for high-frequency rectifiers, for flyback catching in switchers, and anywhere the 0.7 V toll wastes precious headroom. The honest trade-offs: a Schottky leaks more in reverse and has a lower breakdown voltage, so it is not a drop-in everywhere.
Protection is the diode's other quiet career. A single diode in series at a board's power input blocks current if someone plugs the battery in backwards — reverse-polarity protection that costs one forward drop (use a Schottky, or a MOSFET, to claw back that loss). And a diode reversed across a sensitive input shunts any negative spike to ground before it can do harm. Notice the recurring pattern: the diode simply gives stray current a safe one-way path to somewhere it can do no damage.
That closes the diode rung. You now own the whole family: the one-way valve and its 0.7 V model, the rectifier, the reservoir cap and its ripple, the Zener reference — and now the LED, the flyback diode, clippers, clamps, and the Schottky. Every one is the same simple idea pointed at a different problem. Next the ladder climbs to the part that doesn't just steer current but amplifies it: the transistor, the device that turns these passive tricks into active circuits.