Inductor

An inductor is a two-terminal component that stores electrical energy in a magnetic field, built by shaping a conductor so that current flowing through it creates a magnetic field linked to its own circuit. Any change in that current induces a voltage that opposes the change — the inductance, measured in henries, describes how strongly a given rate of current change generates that opposing voltage. Outside a chip, inductors are typically discrete wound-wire components; inside a chip, they're built as spiral or loop traces in the top metal layers, because a coil needs real physical area to generate meaningful magnetic field strength, unlike a resistor or capacitor that shrinks cleanly with geometry. How an inductor actually gets built and used, from raw structure to a finished circuit role, looks like this:

Conductor shaped into a loop or spiral
On-chip: top metal layers; off-chip: wound wire or PCB trace
Inductor formation and behaviorInductance set by loop geometry and surrounding material
Current flows, magnetic field builds
Field strength grows with number of turns and loop area
Changing current induces opposing voltage
Inductor resists sudden current changes
Core material, if present, concentrates field
Higher permeability core raises inductance for the same geometry
↻ next switching cycle
Deployed for power conversion, filtering, or RF tuning
Same physics, different circuit role

A few things are worth knowing about why an inductor is the odd one out among passive components, and why it's so much harder to put on a chip:

An inductor's value grows with loop area and the square of the number of turns, which is exactly why it doesn't shrink the way a resistor or capacitor does when a process node gets smaller. A resistor's value comes from a length-to-width ratio and a capacitor's from plate area over spacing, both of which can shrink in proportion as transistors shrink. An inductor's magnetic field strength depends on physical loop area and how many times a conductor wraps around that area, so shrinking the geometry directly shrinks the inductance — there's no equivalent trick of thinning a dielectric to claw the value back. This is the core reason on-chip inductors are large, inefficient component by chip standards, occupying far more area per unit of electrical function than a transistor, resistor, or capacitor ever would.

Three Ways to Build an Inductor loop geometry sets the field, and the field is what shrinks worst with scaling On-Chip Spiral spiral trace in top metal layers small inductance, large die area, used in RF and on-die power PCB / Package Trace planar trace routed on board or package more area available than silicon, moderate inductance Discrete Wound Inductor wire wound around a ferrite or air core largest inductance per part, used for board-level power supplies same physics at every scale: more turns and more loop area means more stored magnetic energy

An inductor resists sudden current changes but passes a steady current freely — the exact mirror image of a capacitor, which resists voltage changes but passes current freely as voltage changes. This complementary behavior is why the two are paired in nearly every power and filtering circuit: a capacitor smooths voltage, an inductor smooths current, and switching power converters use both together to turn a chopped, switching voltage into a clean, steady output. The energy an inductor stores doesn't disappear when current stops instantly either — it has to go somewhere, which is why an inductor driven by a switch that opens abruptly produces a voltage spike, a behavior that circuit designers either harness deliberately or have to suppress with a protective diode.

A switching voltage regulator is the single most common reason a real inductor shows up on a modern board, and its job is converting one steady voltage into another without wasting the difference as heat. A linear regulator drops excess voltage as heat, which is simple but wastes power proportional to the voltage difference; a switching regulator instead turns a transistor rapidly on and off, and an inductor smooths that chopped square wave into a steady output current at a different voltage, storing energy during the on phase and releasing it during the off phase. Because efficiency matters enormously in battery-powered devices and in powering billions of transistors on a modern AI accelerator, switching regulators with inductors — not linear regulators — are what actually deliver power to a high-performance chip's silicon.

RF circuits use inductors specifically for the one property neither a resistor nor a capacitor alone can provide: a reactance that pushes against a capacitor's reactance to create a sharp resonance at a chosen frequency. An inductor and a capacitor wired together form a resonant circuit that strongly favors one frequency and rejects others, which is the basic mechanism behind radio tuning, oscillators, and the matching networks that connect a chip's RF transistors to an antenna efficiently. Because on-chip inductors are large and lossy compared to their discrete counterparts, RF designers constantly trade off between integrating an inductor on-die for compactness and using an external discrete inductor for better performance — a tradeoff that resistors and capacitors rarely force to this degree.

Read an inductor as stored magnetic energy with a scaling problem, not a shrinking component: unlike a resistor or capacitor, its value is tied directly to physical loop area and turn count rather than a ratio that gets smaller for free as a process node advances, which is why it remains the largest, most reluctantly integrated passive component on any chip, and why its role — smoothing current in a power converter or forming resonance in an RF tank — is usually worth that area cost only when nothing else can do the job.

Take inductor further

Ask the copilot about this term, or have our engineers assess it against your process.