Optocoupler

An optocoupler carries a digital or low-bandwidth analog signal from one electrical domain to another using light, with no metallic path between the two sides. That gap is the whole point — it isolates ground loops, blocks high-voltage transients, and protects fragile logic from noisy power.

On the input side, current through an LED produces light. On the output side, a phototransistor (or photo-darlington, or photo-triac) sees that light and conducts. A clear-plastic dielectric or air gap of typically 0.4 mm separates the two halves, rated for several kilovolts of isolation. Because only photons cross the gap, the input and output grounds can sit at vastly different potentials without leaking current to each other.

In plain terms

Two strangers in soundproof rooms passing notes by flashlight through a window. The light gets through; nothing else does.

Also called: opto-isolator, photocoupler, opto.

Why designers use it

Best for

Key specifications

When not to use it

Common mistakes

Where you will find it

A short history

The optocoupler combines two earlier inventions: the LED (Holonyak, 1962) and the silicon phototransistor (Shive, 1950). The first commercial optocoupler — the General Electric H11A1, packaging a GaAs IR LED with a silicon phototransistor in a 6-pin DIP — appeared in 1968, and the basic four-pin variant followed in 1971 as the standard general-purpose isolator. Sharp introduced the PC817 in the late 1980s, which became the de facto cost-driven workhorse for SMPS feedback loops and remains in production at billion-unit annual volume. Modern silicon-photonic and capacitive digital isolators have largely replaced optocouplers in fast data paths, but the LED + phototransistor optocoupler remains dominant for AC-mains power-supply feedback because it tolerates large common-mode transients and its isolation barrier is well-understood by safety regulators (UL 1577, IEC 60747-5-5).

Good to know

Related