Darlington Transistor: Big Gain, Bigger Personality

1) What a Darlington Transistor Is (and Isn’t)

A darlington transistor—also called a Darlington pair—is a compound BJT where the emitter of the first transistor drives the base of the second. The result is a composite device that behaves like a single, very high-gain transistor. The darlington transistor shows up in relay and solenoid drivers, small DC motor drivers, lamp strings, and any place a microcontroller needs hefty current without a hefty base drive.

But the darlington transistor isn’t a free lunch. You stack base-emitter junctions (raising V_BE(on)), you add saturation voltage, and you slow things down due to charge storage. Used wisely, it’s a star; used blindly, it turns your board into a slow, warm nightlight.

2) Math of a Darlington Transistor: Gain, VBE, and VCE(sat)

2.1 Overall Current Gain

For two BJTs with gains β₁ and β₂, the darlington transistor gain is:

β_total ≈ β₁·β₂ + β₁ + β₂  ≈  β₁·β₂  (when β ≫ 1)

If each transistor has β ≈ 100, the darlington transistor ends up around 10,000—meaning base current can be tiny for a given collector current. That’s why a darlington transistor is beloved in microcontroller land.

2.2 Stacked V_BE

Two silicon junctions in series yield V_BE(on) ≈ 1.2–1.4 V (vs ~0.7 V for one BJT). Your logic pin must be able to forward-bias the darlington transistor despite this higher threshold—still easy from 3.3–5 V GPIO, but it changes resistor math.

2.3 Saturation Voltage

The darlington transistor often shows V_CE(sat) of ~0.8–2.0 V at meaningful current; a single BJT might be ~0.05–0.3 V in similar conditions. That extra volt or so becomes heat (P ≈ I × V_CE(sat)) and voltage lost across the switch—essential in low-voltage loads.

2.4 Base Resistor Sizing (Switching)

// Example: 5 V MCU → darlington transistor → 300 mA solenoid (with diode)
Target I_C = 0.3 A
Choose I_B_total ≈ I_C / 100 (Darlington allows tiny I_B; use margin)
I_B_total ≈ 3 mA
Assume V_BE_total ≈ 1.2 V → R_B ≈ (V_IO − V_BE_total)/I_B_total
R_B ≈ (5 − 1.2)/0.003 ≈ 1267 Ω → pick 1.2 kΩ or 1.3 kΩ

Add a base pulldown (e.g., 100 kΩ) if your input source floats at reset so the darlington transistor stays off.

3) Speed Limits: Storage Time and Why Darlingtons Feel “Sticky”

The darlington transistor tends to store charge during saturation; when you turn it off, that stored charge must leave before current fully stops. The observed effect is a longer storage and fall time compared to a single BJT, and much slower than a MOSFET. In low-frequency applications (relays, solenoids, lamps), who cares. In PWM motor control or audio, this “stickiness” can mean heat and distortion.

• Mitigation: Avoid deep saturation by not over-driving the base excessively; add speed-up networks across base resistor; prefer arrays with built-in shaping.
• Topology: If you need kHz-range hard switching with low loss, the darlington transistor is probably not the lead. Use a MOSFET.

4) Design Recipes: Relays, Solenoids, and Motors

5) Darlington Arrays: ULN2003/ULN2803 and Friends

Why wire a single darlington transistor when you can have seven or eight in a neat DIP or SOIC? Integrated arrays (like the classic 7-channel and 8-channel open-collector NPN Darlington arrays) bundle base resistors and clamp diodes, so you can drive multiple relays/solenoids with one IC from a microcontroller.

• Open-collector outputs: They sink current; tie the load to +V.
• Input resistors inside: Easy MCU interfacing; check input current spec.
• Clamp diode pin: Some arrays expose a COM pin for the freewheel path—wire it to +V for inductive loads.

6) Layout, EMI, and Flyback: The Non-Negotiables

• Short loops: Keep the coil + diode + darlington transistor loop tight to cut ringing.
• Shared ground sanity: Big load currents should not flow through logic ground traces.
• Diode placement: Put the diode at the load if possible. Its job is to catch the inductive tantrum before it reaches the darlington transistor.
• Snubbers: For fast/longer wires, add RC snubbers to tame residual spikes and EMI.

7) Thermal & SOA: Keeping a Darlington Transistor Cool

The math is simple and ruthless: P ≈ I × V_CE(sat). Because V_CE(sat) is higher, a darlington transistor runs warmer at a given current than a single BJT or MOSFET. Choose packages like TO-220/TO-247 for higher currents, and check the safe operating area (SOA) for pulses—relays and solenoids can surprise you on startup.

• Heatsink plan: If P > ~1 W continuous, plan for thermal pad/compound and a modest sink.
• Duty matters: A 10% duty may be fine bare, but 80% duty might demand a heatsink for the same current.
• Ambient and enclosure: Closed boxes make the darlington transistor sweat. Derate accordingly.

8) Darlington vs Single BJT vs MOSFET: Who Stars When?

9) Reading the Datasheet: Curves That Matter

• V_CE(sat) vs I_C: This sets heat and drop; scale current honestly.
• I_B vs I_C: See how little base current you actually need—then add margin.
• Switching: Storage and fall times define PWM limits for your darlington transistor.
• SOA: Respect pulse limits at your supply and inductance.
• Thermal RθJA/RθJC: Model your enclosure and heatsink; don’t forget ambient.

10) Troubleshooting: The Case of the Toasty Darlington

Symptoms → Likely Causes → Fix

• Driver runs hot: V_CE(sat) × I too high → lower current, upgrade package, or move to MOSFET.
• Relay buzzes: Missing diode or weak drive → add flyback, increase base current; check coil voltage.
• Slow PWM response: Storage time → reduce frequency or switch tech; try speed-up cap across base resistor.
• Logic pin glitch turns on load: Floating input → add pulldown/pullup; shield long traces; add RC at input.
• Random resets nearby: Inductive kick EMI → tighten layout; diode at coil; add snubbers and ground discipline.