How Do Light Emitting Diodes Work? A Bench-Tested Circuit Guide

The Bench Reality: How Do Light Emitting Diodes Work in Practice?

When a customer or a junior tech asks me, "how do light emitting diodes work?", I skip the deep quantum mechanics of electron-hole recombination and point straight at the workbench. At its core, an LED is a specialized P-N junction diode. When you forward-bias it (apply positive voltage to the P-type anode and negative to the N-type cathode), electrons cross the junction and drop into lower energy holes, releasing that energy difference as photons. The specific semiconductor materials (like Gallium Arsenide Phosphide for red, or Indium Gallium Nitride for blue) dictate the bandgap, which directly sets the forward voltage ($V_f$) and the color of the light.

But on the bench, you aren't manipulating bandgaps; you are managing current. An LED is a current-driven device. It doesn't care about the voltage across it as much as it cares about the electrons flowing through it. If you force 5V directly across a 2V red LED, it will draw massive current, overheat the microscopic gold bond wire inside, and pop.

Pinout and Symbol Identification

Before you solder, you must correctly identify the terminals. For the standard 5mm (T-1 3/4) through-hole epoxy package:

• Anode (Positive): The longer leg. Inside the clear epoxy lens, it connects to the smaller metal post.
• Cathode (Negative): The shorter leg. Inside the lens, it connects to the larger, flat-topped metal anvil (which acts as a reflector). The outside rim of the epoxy lens also has a distinct flat spot on the cathode side.

Schematic Symbol: The symbol is a triangle pointing toward a vertical line (the cathode barrier). The triangle represents the anode. Two small arrows point away from the junction, indicating emitted light.

Biasing and Selection: Sizing the Current Limiter

Because an LED's voltage-current curve is highly non-linear, a tiny increase in voltage past the $V_f$ threshold causes a massive spike in current. You must use a current-limiting resistor in series. Here is how to bias and select components for a standard indicator circuit.

Complete Application Circuit: 5V USB Indicator

Let's build a reliable power indicator for a 5V USB-powered Arduino Nano project.

1. Source Voltage ($V_s$): 5.0V (from the Nano's 5V pin).
2. Target LED: Standard 5mm Red (Nominal $V_f$ = 2.0V, Target $I_f$ = 15mA for comfortable indoor brightness).
3. Resistor Calculation: Using Ohm's Law: $R = (V_s - V_f) / I_f$.
   $R = (5.0V - 2.0V) / 0.015A = 200\Omega$.
4. Standard Value Selection: 200Ω is not a standard E12 value. We step up to the next standard value: 220Ω. This slightly reduces current to ~13.6mA, extending the LED's lifespan.
5. Resistor Wattage: $P = I^2 \times R = (0.0136)^2 \times 220 = 0.04W$. A standard 1/4W (0.25W) carbon film resistor is more than sufficient and will run completely cool.

Wiring Sequence: 5V Pin → 220Ω Resistor → LED Anode (Long Leg) → LED Cathode (Short Leg) → GND Pin.

Operation Regions and Safe Default Part Numbers

Understanding the three operational regions of the diode junction prevents catastrophic design flaws. Below is the operational matrix for standard 5mm indicator LEDs.

Safe Default Part Numbers (2026 Bench Standards)

When ordering for the shop, stick to high-bin, reputable manufacturers. Cheap, unbranded LEDs from bulk marketplaces often suffer from severe lumen depreciation and wide $V_f$ variances.

• Standard Red: Kingbright WP7113SURCK (2.0V $V_f$, 20mA, 1000mcd, water-clear lens). Typically $0.12/ea in bulk.
• Standard Blue: Lite-On LTL-4233N (3.2V $V_f$, 20mA, 1500mcd). Typically $0.15/ea.
• High-Brightness White: Cree C503B-WAN (3.2V $V_f$, 20mA, 15,000mcd). Use this when the LED must be visible in direct sunlight. Typically $0.35/ea.

For comprehensive component data, always verify your specific batch against the manufacturer's datasheet or trusted educational repositories like the SparkFun LED Tutorial or All About Circuits semiconductor text.

War Story: The Parallel LED Thermal Runaway

Theory is clean; the bench is messy. Here is a real-world scenario that demonstrates why understanding LED physics matters when scaling up a design.

The Setup

A client needed a custom control panel with three white indicator LEDs wired in parallel off a single 5V rail. To save board space, the junior designer used one shared current-limiting resistor for all three LEDs. The white LEDs had a nominal $V_f$ of 3.2V and a target $I_f$ of 20mA each (60mA total).

They calculated the resistor: $R = (5.0V - 3.2V) / 0.060A = 30\Omega$. They installed a standard 33Ω 1/2W resistor.

The Numbers and Outcome

When powered on, the circuit drew exactly 54mA. All three LEDs lit up brightly. The designer signed off on the prototype. However, during a 4-hour burn-in test inside the enclosed panel, the ambient temperature rose to 45°C.

At the 3-hour mark, LED #2 flickered and died. Two minutes later, LED #1 popped audibly and went dark. LED #3 survived but was visibly dimmed and discolored.

What Went Wrong: Thermal Runaway

LEDs have a negative temperature coefficient for forward voltage. As an LED gets hot, its $V_f$ drops. Because the three LEDs were in parallel, they shared the same voltage node, but their individual $V_f$ thresholds were slightly different due to manufacturing tolerances.

1. LED #2 was physically closest to a voltage regulator and got slightly warmer.
2. Its $V_f$ dropped from 3.2V to 3.1V.
3. Because its resistance dropped, it began hogging more of the 54mA total current, pulling perhaps 25mA while the others pulled 14mA each.
4. More current = more heat = lower $V_f$ = even more current. This positive feedback loop is thermal runaway.
5. LED #2 exceeded its 30mA absolute max rating, overheated, and its internal gold bond wire melted (open circuit).
6. With LED #2 dead, the remaining 54mA was now split between only two LEDs (27mA each). They rapidly entered their own thermal runaway cycles and failed in succession.

Troubleshooting: How It Fails and How to Test It

When an LED circuit goes dark, the LED is usually the victim, not the culprit. But you still need to verify it. Do not use the resistance (Ohms) setting on your multimeter; the test voltage is often too low to forward-bias the junction, leading to false "open" readings.

How to Test with a Multimeter

1. Set your digital multimeter (DMM) to the Diode Test mode (usually indicated by a triangle with a line and a soundwave symbol).
2. Connect the Red probe to the LED Anode (long leg) and the Black probe to the Cathode (short leg).
3. Expected Reading: The DMM will output a small test current (usually 1-2mA). The LED should glow faintly. The screen will display the forward voltage drop (e.g., 1.6V to 2.0V for red, 2.8V to 3.2V for blue/white).
4. Reverse the probes (Black to Anode, Red to Cathode). The screen should read "OL" (Over Limit) or "1", indicating the junction is successfully blocking reverse current.

Common Failure Modes

• Reads "OL" in both directions: The LED is dead. The internal bond wire has melted due to overcurrent, or the die has cracked from mechanical stress (bending the legs too close to the epoxy base). Replace it.
• Reads 0.00V or a dead short in both directions: The junction has suffered reverse-breakdown avalanche or extreme thermal damage, melting the semiconductor into a conductive slug. Check your circuit for reverse-voltage spikes or missing current limiters before installing a replacement.
• Reads correct $V_f$ but emits no light: The die is intact, but the phosphor coating has degraded, or the epoxy lens has yellowed from prolonged UV/heat exposure (common in cheap white LEDs). It's electrically functional but optically dead.

Understanding how light emitting diodes work isn't just about knowing that they emit photons. It's about respecting the P-N junction's thermal and electrical limits. By sizing your resistors correctly, avoiding parallel thermal traps, and testing with the right DMM settings, you'll build indicator circuits that outlive the devices they are mounted in.