The bipolar junction transistor
A bipolar junction transistor (BJT) is a sandwich of three doped layers — NPN or PNP — forming two PN junctions back to back. It is called "bipolar" because both electrons and holes take part in conduction. The three regions are not alike:
- Emitter — heavily doped, so it can inject plenty of carriers.
- Base — very thin and lightly doped, so carriers cross it before they can recombine.
- Collector — moderately doped and physically the largest, because it dissipates most of the heat.
In the symbol the arrow is always on the emitter and shows the direction of conventional current. An NPN arrow points out ("Not Pointing iN"); a PNP arrow points in ("Points iN"). Everything about a PNP is the mirror image of an NPN: its supply polarity, bias voltages and currents are all reversed.
How it works: a small current controls a large one
For normal (amplifying) operation:
- the base–emitter junction is forward biased — about 0.6–0.7 V for silicon; and
- the collector–base junction is reverse biased.
The forward-biased emitter junction injects carriers (electrons, in an NPN) into the base. Because the base is so thin, about 95–99% of them diffuse straight across it and are swept into the collector by the reverse-biased collector junction. Only a few recombine and leave through the base lead. The base current is therefore small, and it controls a much larger collector current.
| Relationship | Meaning |
|---|---|
| IE = IB + IC | The emitter carries the base and collector currents together. |
| β (hFE) = IC ÷ IB | DC current gain in common emitter — typically 20–100 for power transistors and 100–500 for small-signal types. |
| α = IC ÷ IE | The fraction of emitter current that reaches the collector — just under 1 (0.95–0.99). |
For example, a base current of 40 µA in a transistor with β = 150 gives a collector current of 6 mA, and an emitter current of 6.04 mA.
Three regions: amplifier or switch
- Cut-off — base-emitter voltage below about 0.5 V, no base current, so no collector current (only tiny leakage). Both junctions are effectively reverse biased, and the transistor is an open switch.
- Active (linear) — collector current is proportional to base current (IC = β IB). This is the amplifier region.
- Saturation — the base is driven so hard that the collector current is limited only by the external circuit. Both junctions are forward biased and VCE(sat) is only about 0.1–0.3 V, so the transistor is a closed switch.
Used as a switch, a transistor is driven between cut-off and saturation and spends almost no time in between, so it wastes very little power. Transistor switches drive relays, lamps, solenoids and LEDs from logic signals. A flyback diode across any inductive load protects the transistor from the voltage spike at switch-off, which could otherwise exceed VCEO.
Characteristics and ratings
| Rating | What it limits |
|---|---|
| VCEO | Maximum collector–emitter voltage (with the base open) |
| VCBO, VEBO | Maximum collector–base and emitter–base reverse voltages. VEBO is small, typically only about 5–7 V. |
| IC max | Maximum continuous collector current |
| Ptot | Maximum power dissipation, ≈ VCE × IC, at a stated case temperature; derated when hot |
| Tj max | Maximum junction temperature (≈ 150–200 °C for silicon) |
| hFE, fT | Current gain (which varies with current and temperature), and transition frequency, where the gain falls to 1 |
Collector leakage current roughly doubles for every 10 °C rise. In a badly designed or poorly cooled power stage, the extra current heats the junction, which raises the current further, until the transistor destroys itself. The protections are emitter (stabilising) resistors, correct bias design, and heat sinks with thermal compound — plus insulating mica washers where the case must be isolated, because on many power transistors the case is the collector.
Component description and orientation
Fitting a transistor the wrong way round, or confusing its leads, can destroy it and the circuit around it. Common packages:
- TO-92 — small plastic package with a flat face. Lead order (E-B-C, C-B-E, …) varies between part numbers.
- TO-18 / TO-39 — small metal cans. A tab on the rim marks the lead next to it (usually the emitter), and the case may be connected to the collector.
- TO-220 — power package whose metal tab, normally the collector, bolts to a heat sink.
- TO-3 — large diamond-shaped metal can with only two pins, base and emitter. The case is the collector.
- SOT-23 and similar surface-mount packages — tiny, soldered directly to the board.
Type numbers follow the same schemes as diodes: 2N (JEDEC: 2N2222, 2N3055), or Pro Electron letters where the second letter gives the use — BC small-signal audio frequency, BD power, BF radio frequency. The data sheet or the component maintenance manual is the final authority.
Other transistor types
- Field-effect transistor (FET) — a unipolar device (only one carrier type) with terminals gate, drain and source. The gate voltage controls the drain current, and almost no gate current flows, so the input impedance is very high. Compare the BJT, which is controlled by base current.
- JFET (junction FET) — N-channel or P-channel. A reverse-biased gate junction narrows the channel; at the pinch-off voltage the channel closes. A JFET is normally on (depletion mode). In the symbol the gate arrow points in for an N-channel.
- MOSFET (metal-oxide-semiconductor FET) — the gate is insulated from the channel by a thin silicon dioxide layer, giving an input resistance of 10¹² Ω or more. Enhancement MOSFETs (broken channel line in the symbol) are normally off and turn on when the gate-source voltage exceeds a threshold. Depletion MOSFETs are normally on. Power MOSFETs switch large currents with very low on-resistance — in solid-state power controllers, motor drives and switch-mode supplies.
- CMOS (complementary MOS) — pairs of N-channel and P-channel MOSFETs, one of which is always off. They draw almost no current except when switching, so power consumption is very low. Most logic chips, microprocessors and memories are CMOS.
- IGBT (insulated-gate bipolar transistor) — a MOSFET-style insulated gate driving a bipolar output stage. It switches high voltages and currents in inverters and motor drives.
- Darlington pair — two transistors in cascade, the first feeding the second's base. The overall gain is about β1 × β2, and VBE is about 1.2–1.4 V.
- Unijunction transistor (UJT) — one junction and three terminals: emitter, base 1 and base 2. It fires suddenly when the emitter voltage reaches a set fraction of the inter-base voltage, which suits relaxation oscillators and SCR trigger circuits.
- Phototransistor — light provides the base current (Chapter 3).
The gate oxide of a MOSFET or CMOS chip is so thin that a static discharge far too small to feel can puncture it. Handle them only with ESD precautions — a grounded wrist strap, a conductive mat and shielding bags (Chapter 7).
B2 extension: configurations and testing
B1 needs only a familiarisation with transistors, but two things are worth knowing because they come up in practice:
- Configurations — the transistor terminal shared by input and output names the circuit. Common emitter gives voltage and current gain with a 180° phase inversion, and is the workhorse amplifier. Common collector (the emitter follower) gives a voltage gain just under 1, high input impedance and low output impedance, so it is used as a buffer. Common base gives voltage gain but no current gain (α < 1) and has a low input impedance; it is used at high frequencies.
- Testing with a multimeter — on the diode-test range, a transistor looks like two diodes sharing the base. For an NPN, the red lead on the base and the black on the emitter, then on the collector, each reads about 0.6–0.7 V, and the reverse reads OL. For a PNP, the black lead goes on the base. Collector–emitter should read OL both ways. A low reading anywhere suggests a short or leakage.