The unijunction transistor (UJT) is a three-terminal semiconductor switch characterized by its unique negative resistance region. While largely superseded by microcontrollers and 555 timers in new commercial designs, the UJT remains a staple in legacy industrial equipment, SCR/triac triggering circuits, and educational bench setups. If you are repairing an older motor drive or building a rugged, low-component-count relaxation oscillator, understanding how to bias, test, and substitute a UJT is essential.

UJT Pinout, Symbol, and the Negative Resistance Concept

Unlike a standard bipolar junction transistor (BJT) that amplifies current, a UJT acts as a voltage-controlled switch. It consists of a lightly doped N-type silicon bar with ohmic contacts at both ends, and a heavily doped P-type emitter alloyed into one side.

Pinout Identification (TO-92 Package): Holding the flat face of a standard 2N2646 TO-92 package toward you with the leads pointing down, the pins from left to right are typically Base 1 (B1), Base 2 (B2), and the Emitter (E). Always verify with a datasheet, as European PRO Electron parts (like the BRY39) may differ.

The Schematic Symbol: The symbol features a straight diagonal line representing the N-type base bar. B1 and B2 connect to the ends of this bar. The Emitter (E) is represented by an arrow pointing inward at an angle toward the bar, indicating the P-N junction.

The core operating principle relies on the intrinsic standoff ratio ($\eta$), a fixed internal voltage divider determined by the physical placement of the emitter on the silicon bar. $\eta$ typically ranges from 0.5 to 0.8. The emitter will not conduct until the applied emitter voltage ($V_E$) exceeds the peak point voltage ($V_P$), calculated as:

$V_P = (\eta \times V_{BB}) + V_D$

Where $V_{BB}$ is the voltage across B2-B1, and $V_D$ is the forward diode drop (approx. 0.7V). Once $V_E$ hits $V_P$, the device fires, dumping stored charge and exhibiting negative resistance—meaning current increases while voltage across the device simultaneously drops.

Operation Regions and Electrical Characteristics

To design with a UJT, you must understand its three distinct operating regions. The table below maps these regions to typical bench measurements for a standard 12V circuit using a 2N2646 ($\eta \approx 0.65$).

Operating Region Emitter Voltage ($V_E$) Emitter Current ($I_E$) Device State & Bench Behavior
Cutoff $V_E < V_P$ (e.g., 0V to 8.4V) Near zero (leakage only, $<1\mu A$) High impedance. The P-N junction is reverse-biased. The timing capacitor is charging.
Negative Resistance $V_P$ down to $V_V$ (e.g., 8.5V dropping to 1.5V) Rapidly increasing (mA range) Device "fires." Holes inject into the N-bar, lowering B1 resistance. Current surges while $V_E$ collapses.
Saturation $V_E < V_V$ (e.g., $<1.5V$) Limited by external circuitry Low impedance ON state. The UJT acts like a closed switch until $I_E$ drops below the valley current ($I_V$).

Note: $V_V$ is the Valley Voltage, and $I_V$ is the Valley Current. If your external charging resistor cannot supply at least $I_V$, the UJT will latch up and fail to oscillate.

Designing a UJT Relaxation Oscillator Circuit

The most common application for a unijunction transistor UJT is the relaxation oscillator, used to generate sawtooth waveforms or trigger pulses for SCRs. Let us design a 1kHz oscillator to trigger a triac in a phase-control circuit.

Target Specifications:

  • Supply Voltage ($V_{BB}$): 12V DC
  • Target Frequency ($f$): ~1,000 Hz (Period $T = 1ms$)
  • UJT: 2N2646 (Assume $\eta = 0.65$)

Component Selection & Biasing Steps:

  1. Calculate Peak Voltage ($V_P$): $V_P = (0.65 \times 12V) + 0.7V = 8.5V$. The capacitor must charge to 8.5V to trigger the UJT.
  2. Select Timing Capacitor ($C_T$): Choose $C_T = 100nF$ ($0.1\mu F$). This is small enough to charge quickly but large enough to bypass stray capacitance.
  3. Calculate Charging Resistor ($R_T$): The timing formula is $T = R_T \times C_T \times \ln(1 / (1 - \eta))$.
    $\ln(1 / 0.35) \approx 1.05$.
    $R_T = 1ms / (100nF \times 1.05) = 9.52k\Omega$. Select standard 10k$\Omega$. This yields a real-world frequency of roughly 950 Hz.
  4. Select Base 1 Resistor ($R_{B1}$): This resistor develops the output trigger pulse. Use 100$\Omega$. A lower value yields a sharper, higher-current pulse but reduces the voltage amplitude.
  5. Select Base 2 Resistor ($R_{B2}$): Used for temperature compensation. As temperature rises, the interbase resistance drops, which would alter $V_P$. Adding a 470$\Omega$ resistor in series with B2 stabilizes the firing point.
Circuit Wiring Summary: Connect $R_T$ (10k) from +12V to the Emitter. Connect $C_T$ (100nF) from Emitter to Ground. Connect $R_{B2}$ (470$\Omega$) from +12V to B2. Connect $R_{B1}$ (100$\Omega$) from B1 to Ground. The output trigger pulse is taken across $R_{B1}$.

Bench Testing: How to Verify a UJT with a Multimeter

UJTs typically fail due to overcurrent conditions that melt the internal emitter junction, resulting in a short between the Emitter and Base 1. Here is how to test a suspected 2N2646 on the bench using a standard digital multimeter (DMM). For deeper component theory, refer to the Electronics Tutorials UJT guide.

  1. Test Interbase Resistance ($R_{BB}$): Set your DMM to the Ohms ($\Omega$) range. Measure between B1 and B2. A healthy 2N2646 will read between 4.0k$\Omega$ and 9.1k$\Omega$ (check your specific datasheet). If it reads 0$\Omega$ (short) or OL (open), the device is dead.
  2. Test Emitter Junction (Forward Bias): Switch the DMM to Diode Test mode. Place the Red probe on the Emitter (P-type) and the Black probe on B1, then B2. You should read a standard silicon diode drop: 0.450V to 0.750V.
  3. Test Emitter Junction (Reverse Bias): Swap the probes (Black on Emitter, Red on B1/B2). The meter should read OL (Over Limit / Open). If it reads a low voltage or continuity, the P-N junction is shorted.

Safe Default Part Numbers and Modern Alternatives

If you are repairing legacy gear or stocking your lab, these are the benchmark part numbers:

  • 2N2646: The undisputed hobbyist and educational standard. N-type, 30V max $V_{B2B1}$, 50mA max emitter current, $R_{BB}$ 4-9k$\Omega$. Read the 2N2646 datasheet specs on Components101 for exact thermal derating curves.
  • 2N2160: A heavier-duty alternative, often found in older industrial SCR gate drives. Rated for higher peak pulse currents.
  • The Modern Substitute (PUT): True UJTs are largely obsolete in new surface-mount production. If you cannot source a 2N2646, use a Programmable Unijunction Transistor (PUT) like the 2N6027. A PUT is actually a 4-layer thyristor, but by using two external resistors to set the gate voltage, you can program the exact $\eta$ ratio and mimic UJT behavior perfectly in a relaxation oscillator.

Unijunction Transistor UJT FAQ

Can I substitute a 555 timer for a unijunction transistor UJT?

Yes, in almost all new designs. A 555 timer configured in astable mode can replicate the sawtooth and pulse outputs of a UJT relaxation oscillator with much better frequency stability and wider duty-cycle control. However, a 555 requires more pins, a larger PCB footprint, and bypass capacitors. In high-noise industrial environments (like welding equipment or heavy motor drives), the simple, rugged, two-resistor UJT circuit is sometimes preferred for its immunity to false triggering and lower component count.

Why does my UJT relaxation oscillator frequency drift with temperature?

Frequency drift occurs because the interbase resistance ($R_{BB}$) of the silicon bar has a positive temperature coefficient, while the emitter diode drop ($V_D$) has a negative temperature coefficient (dropping ~2mV/°C). If you omit the Base 2 temperature compensation resistor ($R_{B2}$), the shifting voltage divider alters the peak firing voltage ($V_P$), causing the capacitor to charge to a different threshold. Always include an $R_{B2}$ resistor (typically 10% to 20% of the $R_{BB}$ value, such as 470$\Omega$) to stabilize the firing point across thermal ranges.

What is the difference between a UJT and a Programmable UJT (PUT)?

Internally, they are completely different devices. A true UJT (like the 2N2646) is a three-terminal device with a single P-N junction and a fixed intrinsic standoff ratio ($\eta$) set during manufacturing. A PUT (like the 2N6027) is a four-layer, three-terminal PNPN thyristor (similar to an SCR) with an Anode, Cathode, and Gate. The PUT earns the "programmable" name because you use an external resistor divider on the Gate pin to manually set the trigger voltage, effectively allowing you to dial in any $\eta$ ratio you want. In 2026, PUTs are vastly more available and cheaper than true UJTs.

How do I prevent a UJT from latching up in a DC circuit?

Latch-up happens when the charging resistor ($R_T$) is too small, allowing it to supply more current than the UJT's Valley Current ($I_V$) after the device fires. If $I_E$ never drops below $I_V$, the UJT stays in the low-impedance saturation region and refuses to turn off, halting oscillation. To fix this, ensure your charging resistor is large enough to limit the maximum current below the $I_V$ threshold (typically 2mA to 5mA for a 2N2646). For a 12V supply, an $R_T$ of at least 3k$\Omega$ is usually required to guarantee turn-off.