Zener diodes
An ordinary diode must never be driven into reverse breakdown. A Zener diode is made to work there. Its doping is set so that breakdown happens at a precise voltage, the Zener voltage VZ, anywhere from about 2.4 V to 200 V. Past the knee, the voltage across it stays almost constant while its current changes over a wide range. In the forward direction it behaves like any silicon diode, dropping about 0.7 V.
- Connection: reverse biased, cathode to the more positive point, always with something to limit the current. In the classic shunt regulator it sits in parallel with the load with a series resistor RS. If the input voltage rises or the load current falls, the Zener takes more current, RS drops the extra voltage, and the output stays at VZ. Size the resistor with RS = (Vin − VZ) ÷ (IZ + IL).
- Power rating: PZ = VZ × IZ. For example, a 12 V Zener carrying 50 mA dissipates 0.6 W. The maximum Zener current, IZ max, follows from the rating, and the minimum, IZ min, is just past the knee.
- Temperature coefficient: below about 5 V the true Zener effect dominates and VZ falls slightly as temperature rises (a negative coefficient). Above about 6 V avalanche dominates and VZ rises (positive). Diodes near 5–6 V, and special reference diodes, are almost temperature-stable.
- Uses: voltage references in power supplies and measuring circuits, simple regulators for small loads, clamping and overvoltage sensing — for example in voltage regulators and protection units.
Signal, power rectifier and Schottky diodes
- Signal diodes are small, glass- or plastic-bodied junctions for currents up to a few hundred milliamps, with fast switching (a 1N4148 recovers in a few nanoseconds). They are used for logic, detection, switching and protection.
- Power rectifier diodes have large junctions, high current and PIV ratings, and stud or press-fit packages for heat sinks (Chapter 2).
- Schottky diodes replace the PN junction with a metal-to-semiconductor junction (metal on N-type silicon). Only majority carriers take part, so there is no stored charge to remove. They therefore switch extremely fast and have a low forward drop of about 0.2–0.4 V, which wastes less power. The price is higher leakage and generally lower reverse-voltage ratings. They are used in switch-mode power supplies, high-frequency rectifiers, RF detectors and mixers, low-loss "OR-ing" diodes that combine power sources, and Schottky TTL logic.
Light-emitting diodes
When an electron in a forward-biased junction falls into a hole, it gives up energy. In silicon that energy becomes heat. In compound semiconductors such as gallium arsenide, gallium phosphide, aluminium-gallium-indium phosphide and indium-gallium nitride it comes out as light (electroluminescence). The colour depends on the material's band gap, which also sets the forward voltage.
| Colour | Typical VF | Typical material |
|---|---|---|
| Infrared | ≈ 1.2–1.5 V | GaAs |
| Red / amber | ≈ 1.8–2.2 V | AlGaInP, GaAsP |
| Green | ≈ 2.0–3.2 V | GaP, InGaN |
| Blue / white | ≈ 3.0–3.6 V | InGaN (white = blue LED + phosphor) |
- An LED is forward biased and its current must be set by a series resistor or a constant-current driver: R = (Vsupply − VF) ÷ IF. Without it the current runs away and the LED burns out.
- LEDs have a low reverse-voltage rating, typically about 5 V. Where reverse voltage can appear, protect them with an anti-parallel diode.
- Advantages: very long life, low power and heat, instant switch-on, no filament to break under vibration, small size. Limitations: they need current control, and light output and life fall at high temperature.
- On aircraft: annunciators and push-button legends, displays and display backlighting, cabin and emergency path lighting, and increasingly every exterior light — navigation, anti-collision, logo, landing and taxi. Seven-segment LED displays come in common-anode and common-cathode types.
A red LED (VF ≈ 2 V, IF = 20 mA) is to be lit from the 28 V bus. R = (28 − 2) ÷ 0.02 = 1 300 Ω. The resistor dissipates I²R = 0.02² × 1 300 ≈ 0.52 W, so use a 1 W part — the series resistor of an LED on a 28 V aircraft supply gets surprisingly warm.
Photodiodes, photoconductive cells and optocouplers
- A photodiode has a window or lens over its junction and is operated reverse biased — the photoconductive mode. Light energy frees electron-hole pairs in and near the depletion region, so the reverse current rises in proportion to the light intensity. With no light, only the small dark current (leakage) flows. Photodiodes respond in nanoseconds, so they are used in fibre-optic and data-link receivers, optical encoders, smoke detectors and ambient-light sensors that dim displays automatically.
- With no bias at all a photodiode generates a small voltage — the photovoltaic mode used by solar cells.
- A photoconductive cell (light-dependent resistor, usually cadmium sulphide) is not a junction device. Its resistance simply falls as light increases, much more slowly than a photodiode responds.
- A phototransistor is a light-sensitive transistor: light provides the base current and the transistor amplifies it.
- An optocoupler (opto-isolator) puts an LED and a photodiode or phototransistor in one sealed package. The signal crosses as light, so the input and output circuits have no electrical connection. It is used to isolate control circuits from high-voltage or noisy circuits, for example inside solid-state power controllers.
Varactors and varistors
- A varactor (varicap) diode is used as a voltage-variable capacitor. It is reverse biased, and its junction capacitance falls as the reverse voltage rises, because the depletion layer widens. A DC tuning voltage can therefore tune oscillators and receivers electronically — voltage-controlled oscillators, frequency synthesisers and automatic frequency control in communication and navigation radios.
- A varistor, or voltage-dependent resistor (VDR), is not a junction diode. It is a disc of metal-oxide grains (zinc oxide; older ones silicon carbide), hence the name MOV. Its resistance is very high at normal voltage and collapses above its clamping voltage, equally in both polarities. Fitted across (in parallel with) a supply or component, it conducts the energy of a transient — a switching spike, an inductive kick or a lightning-induced surge — and clamps the voltage. Each large surge degrades it a little, and it is often paired with a fuse. Varistors also suppress arcing across relay and switch contacts.
Avionics inputs are also commonly protected by transient voltage suppressor (TVS) diodes — avalanche diodes that clamp even faster than a varistor — as part of the equipment's lightning and transient protection.
Thyristors
A thyristor is a four-layer (PNPN) semiconductor switch. Inside, it behaves like a PNP and an NPN transistor connected so that each feeds the other's base. Once one starts conducting, the pair drives itself fully on (regenerative action) and stays on — it latches.
- Shockley diode — a two-terminal PNPN diode. It blocks in the forward direction until the breakover voltage is reached, then switches on and stays on until the current falls below the holding current.
- Silicon controlled rectifier (SCR) — the most common thyristor, with three terminals: anode, cathode and gate (connected to the P layer next to the cathode). With the anode positive, it blocks until a positive gate pulse (gate positive with respect to the cathode) triggers it. It then conducts with only about 1–2 V across it. It latches: the gate loses control, and the SCR stays on until the anode current falls below the holding current (IH). Right after triggering, the current must reach the slightly higher latching current (IL), or it drops out again. In reverse it blocks like a diode.
- Turning an SCR off: reduce the anode current below IH. On AC this happens automatically at the end of every positive half cycle (natural commutation). On DC the circuit must be interrupted, or the current forced to zero by a commutating circuit. A negative gate signal does not turn an ordinary SCR off.
- Unwanted turn-on can be caused by exceeding the forward breakover voltage, a very fast rate of rise of anode voltage (dV/dt), or high temperature. RC snubber networks across the SCR guard against dV/dt.
- Uses: controlled rectifiers and battery chargers, AC phase control of heaters, lamps and motors (fire later in each half cycle and less power reaches the load), static switches, and the crowbar overvoltage protection circuit, where an SCR fires to short-circuit the supply and blow a fuse or trip a breaker before an overvoltage can damage equipment. Solid-state turbine ignition exciters also use SCRs to discharge their storage capacitors.
- Triac — in effect two SCRs connected in inverse parallel with a common gate, with terminals MT1, MT2 and gate. It conducts in both directions once triggered and switches off at each current zero, which makes it the AC power switch for dimmers, heater controls, small motors and solid-state relays.
- Diac — a two-terminal bidirectional trigger diode with no gate. It breaks over at about 30 V in either direction, delivering a sharp pulse — usually to trigger a triac.
Zener: reverse biased. LED: forward biased with a series resistor. Photodiode: reverse biased. Varactor: reverse biased, capacitance falls as voltage rises. SCR: removing the gate signal does not switch it off; only reducing the anode current below the holding current does.