When makers and engineers refer to a transistor diode, they are typically talking about one of two things: a Bipolar Junction Transistor (BJT) intentionally wired to act as a PN junction diode, or the intrinsic parasitic body diode found inside a MOSFET. To use a standard NPN BJT as a diode, you short the base and collector pins together; the base-emitter junction then functions as a diode with a forward voltage drop of roughly 0.65V. The safest, most ubiquitous default part for this is the 2N3904 (NPN, 40V, 200mA) or its PNP complement, the 2N3906.
While a standard 1N4148 signal diode is fine for general rectification, wiring a transistor as a diode unlocks critical advantages in analog design—specifically thermal tracking and precise small-signal clamping. Below, we break down the pinouts, biasing rules, a complete current-mirror application, and exactly how to test these configurations on your bench.
The Diode-Connected BJT: Pinout, Symbol, and Operation Regions
To create a diode-connected BJT, you must physically tie the base and collector together. In a standard TO-92 package (flat side facing you, pins pointing down), the pinout for a 2N3904 is Emitter (E), Base (B), Collector (C). You will place a jumper wire or solder bridge between the middle pin (Base) and the right pin (Collector).
In schematic symbols, this is drawn as a standard NPN or PNP transistor symbol, but with a distinct wire looping from the collector node directly into the base node. The anode of your new 'diode' is the Base/Collector node, and the cathode is the Emitter.
| Bias Condition | Voltage at Base/Collector (vs Emitter) | Current Flow | Physical Behavior |
|---|---|---|---|
| Forward Biased (Active) | +0.60V to +0.75V | 10μA to 100mA | Base-emitter junction conducts. Acts as a standard PN diode with a logarithmic V-I curve. |
| Zero Bias | 0V | 0A | Depletion region is at equilibrium. No current flows. |
| Reverse Biased (Safe) | 0V to -5.0V | < 10nA (Leakage) | Junction blocks current. Safe operating area for standard small-signal BJTs. |
| Reverse Breakdown | < -6.0V (V_EBO) | Uncontrolled | Base-emitter Zener breakdown occurs. Can permanently degrade transistor h_FE gain. |
How to Bias and Select a Transistor Diode for Your Circuit
You cannot connect a diode-connected transistor directly across a voltage source without a current-limiting resistor; it will draw excessive current and destroy the silicon die. Because the base and collector are shorted, the transistor is forced into the edge of saturation, and the base-emitter junction dictates the current limit.
Unlike a standard silicon diode, a diode-connected BJT has a highly predictable small-signal dynamic resistance ($r_e$), calculated as $r_e = V_T / I_C$. At room temperature ($V_T \approx 26mV$) and a bias current of 1mA, the dynamic resistance is exactly 26Ω. This makes it vastly superior to a 1N4148 for precision small-signal clamping and logarithmic amplifiers.
Selection Guide and Safe Defaults
When selecting a part, match the voltage and current ratings to your circuit, but also consider the thermal coupling requirement. Here are the safe defaults for your component bin:
- 2N3904 (NPN) / 2N3906 (PNP): 40V $V_{CEO}$, 200mA $I_C$. Cost: ~$0.05/ea. Best for low-power signal clamping, current mirrors, and V_BE multipliers.
- 2N2222 (NPN) / 2N2907 (PNP): 40V $V_{CEO}$, 800mA $I_C$. Cost: ~$0.10/ea. Use when you need to sink/source up to half an amp, such as in LED driver biasing.
- TIP31 (NPN): 60V $V_{CEO}$, 3A $I_C$. Cost: ~$0.40/ea. Use in TO-220 packages when you need to bolt the 'diode' directly to a heatsink alongside a power pass transistor for thermal tracking.
For deeper reading on BJT junction physics and safe operating areas, refer to the ON Semiconductor 2N3904 datasheet and the All About Circuits BJT textbook chapter.
Application Circuit: Thermal-Compensated Current Mirror
The most common practical use for a transistor diode is the current mirror. In this circuit, the diode-connected transistor (Q1) sets a precise reference voltage, which is fed to the base of a second transistor (Q2). Because both transistors share the same $V_{BE}$ characteristics, Q2 mirrors the current flowing through Q1.
This is heavily used in audio amplifiers, LED drivers, and sensor biasing. If Q1 and Q2 are thermally coupled (e.g., glued together or on the same heatsink), the current remains perfectly stable even as ambient temperature changes, because the -2mV/°C temperature coefficient of both junctions cancels out.
Component List
- Q1: 2N3904 NPN (Diode-connected)
- Q2: 2N3904 NPN (Output transistor)
- R_ref: 10kΩ 1/4W resistor (Reference current setter)
- R_load: 1kΩ 1/4W resistor (Simulated load for testing)
- Power Supply: 12V DC bench supply
Build Steps
- Prepare Q1: Insert the 2N3904 into the breadboard. Bend the Collector (right pin) and Base (middle pin) leads together and solder them, or use a short jumper wire to short them on the breadboard. Connect the Emitter (left pin) to Ground.
- Set the Reference: Connect one end of the 10kΩ $R_{ref}$ resistor to the +12V rail. Connect the other end to the shorted Base/Collector node of Q1. This sets the reference current: $I_{ref} = (12V - 0.65V) / 10,000\Omega = 1.135mA$.
- Wire Q2: Insert the second 2N3904. Connect its Base to Q1's Base. Connect its Emitter to Ground. The Base-Emitter junctions are now in parallel.
- Connect the Load: Connect the 1kΩ $R_{load}$ resistor between the +12V rail and Q2's Collector.
- Verify: Power the circuit. Measure the voltage across $R_{load}$. You should read approximately 1.135V, proving that Q2 is sinking exactly 1.135mA, mirroring Q1 regardless of the load resistance (as long as Q2 remains in the active region and doesn't saturate).
Failure Modes and Multimeter Testing Procedures
Transistor diodes fail differently than standard glass-encapsulated diodes. The most common failure mode is Base-Emitter Reverse Breakdown. The $V_{EBO}$ (Emitter-Base Breakdown Voltage) of a 2N3904 is typically only 6V. If you accidentally apply a reverse voltage greater than 6V, the junction undergoes Zener breakdown. While it may still function as a diode, the high-energy avalanche injects carriers into the base oxide, permanently degrading the transistor's forward current gain ($h_{FE}$).
Another failure mode is Thermal Runaway. If the current-limiting resistor is undersized, the junction heats up, which lowers the $V_{BE}$ requirement, which draws more current, leading to a catastrophic short.
How to Test with a Digital Multimeter (DMM)
Use your DMM's dedicated Diode Test mode (usually indicated by a diode symbol). For a comprehensive guide on multimeter usage, check out the SparkFun Transistor Tutorial.
- Isolate the Component: Remove the transistor from the circuit. Testing in-circuit will yield false readings due to parallel paths.
- Forward Bias Test: Place the DMM's Red probe on the Base/Collector node (Anode) and the Black probe on the Emitter (Cathode). A healthy 2N3904 will display a voltage drop between 0.600V and 0.700V.
- Reverse Bias Test: Swap the probes (Black on Base/Collector, Red on Emitter). The meter should read 'OL' (Over Limit) or '1', indicating an open circuit.
- Collector-Emitter Check: Because the base and collector are shorted in this configuration, testing from Collector to Emitter will yield the exact same readings as Base to Emitter. If you read 0.000V in either direction, the silicon is shorted and the part is dead.
Frequently Asked Questions About Transistor Diodes
Can I use a MOSFET body diode instead of a separate freewheeling diode?
Yes, but with strict caveats. Every power MOSFET has an intrinsic parasitic body diode between the source and drain (cathode at drain for N-channel). It will safely clamp inductive kickback from relays or solenoids in low-frequency (<100Hz) switching applications. However, the body diode has a very slow reverse recovery time ($t_{rr}$), often in the hundreds of nanoseconds. If you are driving a motor or inductive load with high-frequency PWM (e.g., 20kHz+), the slow body diode will cause massive switching losses and voltage ringing. In high-frequency scenarios, you must place an external fast-recovery or Schottky diode in parallel with the MOSFET to bypass the body diode.
Why use a diode-connected transistor instead of a standard 1N4148 signal diode?
The primary reason is thermal tracking. A standard 1N4148 is packaged in glass and sits on the PCB. A diode-connected BJT can be mounted directly onto the same heatsink as the power transistor it is biasing (such as in a Class-AB audio output stage or a linear voltage regulator). Because both silicon dies experience the exact same temperature, their $V_{BE}$ voltage drops shift together at -2mV/°C. This prevents the power transistor from drifting into thermal runaway as it heats up under load. A 1N4148 cannot provide this localized thermal feedback.
What happens if I accidentally reverse-bias the base-emitter junction?
If you reverse-bias the base-emitter junction of a standard small-signal BJT beyond its $V_{EBO}$ rating (usually 5V to 7V), the junction will avalanche and conduct current in reverse. This does not always immediately destroy the part, but it causes microscopic lattice damage and traps charge carriers in the base region. The transistor will still work as a diode, but its $h_{FE}$ (DC current gain) will be permanently and significantly reduced. If you need a transistor diode to block reverse voltages, you must place a standard rectifier diode (like a 1N4007) in series with the emitter to protect the delicate base-emitter junction.
