If you are looking for an amplifier, look elsewhere. The unijunction transistor (UJT) is a specialized, three-terminal semiconductor device designed to do one thing exceptionally well: act as a voltage-controlled switch with a negative resistance region. While largely replaced by the 555 timer and microcontrollers in modern commercial consumer electronics, UJTs remain essential for repairing vintage analog synthesizers, building robust high-temperature pulse generators, and understanding negative-resistance circuit theory on the bench.
The undisputed safe default for 95% of hobbyist and educational applications is the 2N2646. Below, we break down exactly how to read its pinout, bias it into oscillation, build a complete relaxation oscillator, and test a suspect part with a standard digital multimeter (DMM).
UJT Pinout, Symbol, and Core Specifications
Unlike a bipolar junction transistor (BJT) with a Collector, Base, and Emitter, a UJT features an Emitter (E) and two bases: Base 1 (B1) and Base 2 (B2). The schematic symbol looks like a standard NPN transistor, but the arrow on the emitter points toward a solid diagonal bar representing the silicon channel, with B1 and B2 extending from opposite ends of that bar.
In the standard TO-18 metal can package, the protruding tab indicates the Emitter. Looking at the bottom of the component with the tab pointing toward you (12 o'clock), the left pin (roughly 8 o'clock) is Base 1, and the right pin (roughly 4 o'clock) is Base 2. Always verify against the specific manufacturer's datasheet, as plastic TO-92 variants occasionally swap the base positions.
The behavior of a UJT is governed by its internal voltage divider and the intrinsic standoff ratio ($\eta$). Here are the critical datasheet parameters for the most common parts you will encounter, sourced directly from the ON Semiconductor 2N2646 Datasheet:
| Parameter | Symbol | 2N2646 (Standard) | 2N2647 (High $\eta$) | Unit |
|---|---|---|---|---|
| Interbase Resistance ($V_{EB2}=0$, $I_E=0$) | $R_{BB}$ | 4.0k – 9.1k | 4.0k – 9.1k | $\Omega$ |
| Intrinsic Standoff Ratio | $\eta$ | 0.56 – 0.75 | 0.68 – 0.82 | Ratio |
| Peak Point Emitter Current | $I_P$ | 5.0 (max) | 5.0 (max) | $\mu A$ |
| Valley Point Emitter Current | $I_V$ | 3.5 (typ) | 3.5 (typ) | mA |
| Valley Point Voltage ($I_V$ condition) | $V_V$ | 1.0 (typ) | 1.5 (typ) | V |
| Emitter to Base 1 Saturation Voltage | $V_{EB1(sat)}$ | 2.5 (max) | 3.0 (max) | V |
Operation Regions and Biasing Rules
To use a UJT, you must understand its three distinct operating regions. The magic happens in the negative resistance region, where an increase in emitter current causes a decrease in emitter voltage. This is what allows the UJT to discharge a capacitor rapidly, creating the sharp pulses used in thyristor triggering and relaxation oscillators.
| Region | Condition | Behavior | Typical Voltages/Currents |
|---|---|---|---|
| Cut-off | $V_E < V_P$ | Emitter diode is reverse-biased. Only tiny leakage current flows. | $V_E$: 0V to $V_P$ $I_E$: < 5 $\mu A$ |
| Negative Resistance | $V_P > V_E > V_V$ | Emitter diode turns on, injecting holes into the B1 region. $R_{B1}$ drops drastically. Current rises as voltage falls. | $V_E$: Drops from $V_P$ to $V_V$ $I_E$: Rises to $I_V$ |
| Saturation | $V_E \le V_V$ | The device is fully turned on. Acts like a closed switch between E and B1. | $V_E$: ~1.0V to 2.5V $I_E$: Limited by external circuit |
The emitter must reach the peak point voltage to trigger the UJT. Calculate it using the formula: $V_P = \eta V_{BB} + V_D$, where $V_{BB}$ is the voltage across the two bases, and $V_D$ is the internal diode forward voltage drop (typically 0.6V to 0.7V at room temperature). If your $V_{BB}$ is 12V and $\eta$ is 0.65, your UJT will fire when the emitter reaches roughly 8.4V.
How to Select and Bias for Oscillation:
For a relaxation oscillator, the external emitter resistor ($R_E$) must force the load line to intersect the negative resistance region. If $R_E$ is too large, the capacitor charges but never reaches the peak current ($I_P$) required to fire. If $R_E$ is too small, the circuit provides more current than the valley current ($I_V$), and the UJT latches on permanently, failing to oscillate.
- Maximum $R_E$: $R_{E(max)} = (V_{CC} - V_P) / I_P$
- Minimum $R_E$: $R_{E(min)} = (V_{CC} - V_V) / I_V$
Your chosen $R_E$ must fall strictly between these two calculated boundaries.
The Classic 1kHz UJT Relaxation Oscillator
Let’s build a practical, free-running relaxation oscillator. This circuit charges a capacitor through a resistor until the UJT fires, rapidly dumping the capacitor's charge through Base 1, creating a sharp positive pulse across $R_{B1}$ and a negative spike across $R_{B2}$. For deeper circuit topology references, Electronics Tutorials provides excellent supplementary waveform analysis.
Target Specifications: ~1kHz output frequency, 12V DC supply, sharp trigger pulses for driving an SCR or logic gate.
Component Bill of Materials (BOM)
- Q1: 2N2646 Unijunction Transistor
- C1: 100nF (0.1$\mu$F) Ceramic or Film Capacitor (rated 25V+)
- R_E: 10k$\Omega$ Resistor (1/4W) — Timing Resistor
- R_B1: 100$\Omega$ Resistor (1/4W) — Pulse Output Resistor
- R_B2: 470$\Omega$ Resistor (1/4W) — Temperature Compensation Resistor
Wiring and Netlist
- Connect the positive rail (+12V) to one end of $R_E$ (10k$\Omega$).
- Connect the other end of $R_E$ to the Emitter (E) of the 2N2646 and the positive leg of $C1$ (100nF).
- Connect the negative leg of $C1$ to circuit Ground.
- Connect Base 2 (B2) to the +12V rail through $R_{B2}$ (470$\Omega$).
- Connect Base 1 (B1) to Ground through $R_{B1}$ (100$\Omega$).
- Output: Take your trigger pulse from the junction of Base 1 and $R_{B1}$.
Frequency Calculation and Verification
The frequency of a UJT relaxation oscillator is determined by the RC time constant and the intrinsic standoff ratio. The formula is:
$f = \frac{1}{R_E \times C \times \ln(\frac{1}{1 - \eta})}$
Assuming a typical $\eta$ of 0.6 for our 2N2646:
- $\ln(1 / (1 - 0.6)) = \ln(2.5) \approx 0.916$
- $f = 1 / (10,000 \times 0.0000001 \times 0.916)$
- $f = 1 / 0.000916 \approx 1091 \text{ Hz}$
On your oscilloscope, you should see a sawtooth waveform at the Emitter (ramping from ~1.5V up to ~8.4V, then snapping back down) and a series of sharp ~1V positive spikes at Base 1 at roughly 1.09kHz. If you need exactly 1.00kHz, swap the 10k$\Omega$ fixed resistor for a 15k$\Omega$ potentiometer and tune it while monitoring the scope.
Failure Modes and Multimeter Testing
UJTs are rugged, but they are not immune to abuse. The most common failure mode is an Emitter-to-Base short. This usually happens when the capacitor value is too large, or the supply voltage is too high, causing the instantaneous peak emitter current ($I_P$) to exceed the device's absolute maximum rating (typically 2A for a microsecond pulse, but much lower for sustained discharge) during the capacitor dump phase. Thermal runaway is a secondary failure mode if $R_{B2}$ is omitted, as the interbase resistance has a positive temperature coefficient that can shift the peak voltage and cause erratic firing or latching.
You can test a suspect UJT in-circuit (if isolated from low-impedance parallel paths) or out-of-circuit using a standard digital multimeter.
- Set your DMM to Diode Test mode. Place the red probe on the Emitter (E) and the black probe on Base 1 (B1). You should read a forward voltage drop between 0.45V and 0.75V.
- Test Emitter to Base 2. Keep the red probe on E and move the black probe to B2. You should read a similar forward voltage drop (0.45V to 0.75V). If either of these reads 0.00V (short) or OL (open), the internal PN junction is destroyed.
- Reverse bias check. Swap the probes (black on E, red on B1, then red on B2). The meter should read OL (open loop) in both cases, confirming the junctions are not leaking backward.
- Switch your DMM to Resistance (Ohms) mode. Measure across Base 1 and Base 2. The meter should display the interbase resistance ($R_{BB}$). For a 2N2646, this must fall between 4.0k$\Omega$ and 9.1k$\Omega$. If it reads near 0$\Omega$ or infinite, the silicon bar is cracked or melted.
- The 'Turn-On' Check (Advanced). While measuring resistance between E and B1 (red on E, black on B1), momentarily short the Emitter to Base 2 with a jumper wire or your finger (if safe/low voltage). The resistance between E and B1 should drop dramatically, indicating the device has been triggered into conduction.
UJT vs. PUT vs. 555 Timer: Which Should You Use?
While the standard UJT is a fantastic educational tool and a drop-in replacement for vintage gear repairs, modern designs usually favor alternatives. Here is how the standard UJT stacks up against the Programmable Unijunction Transistor (PUT) and the ubiquitous 555 timer IC.
| Feature | Standard UJT (2N2646) | PUT (2N6027) | 555 Timer (NE555) |
|---|---|---|---|
| Internal Structure | Single PN junction, N-type bar | 4-layer PNPN (Thyristor-like) | Complex CMOS/Bipolar IC (20+ transistors) |
| Standoff Ratio ($\eta$) | Fixed by factory doping (0.5 - 0.8) | Programmable via external resistor divider | Fixed internally at 1/3 and 2/3 $V_{CC}$ |
| Timing Accuracy | Poor (Highly temperature dependent) | Moderate | High (with stable external RC and bypass caps) |
| Maximum Frequency | ~100 kHz | ~50 kHz | ~500 kHz (Bipolar), >2 MHz (CMOS variants) |
| Best Application | Vintage synth repair, simple SCR triggers | Adjustable threshold crowbar circuits, custom oscillators | PWM generation, precision timing, astable/monostable logic |
The Verdict: Choose the 2N2646 UJT when you are restoring a 1970s analog keyboard, need a simple, low-component-count SCR phase-angle trigger, or want to demonstrate negative resistance on a breadboard. Choose a PUT (like the 2N6027) if you need to dial in a specific trigger voltage that a fixed UJT cannot provide. For everything else—especially if you need precise duty cycle control, logic-level compatibility, or frequencies above 100kHz—skip the discrete transistors and use a CMOS 555 timer (like the TLC555 or LMC555) or a microcontroller GPIO pin.
