Molded Power Inductors for POL Converters

A point-of-load converter is placed as close as possible to the circuit it feeds. The proximity is not a layout preference; it is a consequence of delivering high current at low voltage, where even a small amount of trace resistance or inductance degrades regulation. The converter therefore lives in the same dense area as the processor or logic it supplies, and its inductor has to work in that environment.

What a POL converter does

A POL converter steps a higher intermediate rail down to the voltage a specific load requires, typically from a distribution bus to a core or I/O voltage in the region of a volt or less. It regulates that output against rapidly changing load current, because the load is usually a digital device whose demand swings with its activity.

The inductor in that converter performs two functions at once: it stores energy during each switching cycle, and it filters the ripple current so that the load sees a stable supply. Both functions have to be satisfied in a footprint barely larger than the component itself.

Why these rails use molded inductors

Molded power inductors are built by encapsulating the winding in a compressed metal-composite body. The construction combines several properties that point-of-load designs need at the same time.

• Low profile, since boards are often height-constrained by adjacent components
• Shielded body that limits radiated field near sensitive traces
• Distributed air gap, so inductance falls gradually rather than sharply under bias
• Surface-mount termination compatible with standard reflow assembly
• Construction that supports flat-wire windings for low direct-current resistance

The DCR constraint

At low output voltage and high current, conduction loss in the inductor is a large share of total loss. Direct-current resistance therefore becomes a primary specification rather than a secondary detail: a part with excellent magnetic properties but high DCR can dissipate more power than the converter can afford to remove from such a small area.

This is why POL inductors often use flat-wire or foil windings, which fill the available window more completely than round wire and reduce resistance for a given cross-section.

Switching frequency and core loss

Higher switching frequency reduces the inductance required for a given ripple, which allows a physically smaller inductor and faster transient response. The trade is that core loss rises with frequency, and at MHz switching the core material contributes significantly to total loss.

The practical consequence is that core loss has to be evaluated at the actual switching frequency and flux swing rather than estimated from a single datasheet point. Two cores with similar nominal specifications can behave quite differently once the ripple waveform at the real operating point is applied.

Transient response

When a digital load changes state, its current demand can move faster than the converter can adjust. The inductor and the output capacitors supply the difference in the interim, so the inductance value directly affects how far the output voltage deviates and how quickly it recovers. Lower inductance improves transient response but increases ripple current, which raises core loss and requires more output capacitance.

That trade is the central design decision in a POL stage, and it is resolved with the inductor and the capacitor network considered together rather than separately.

Engineering requirements to define

• Input and output voltage of the rail
• Output current at steady state and worst-case transient
• Switching frequency of the converter
• Target ripple current and resulting inductance
• Saturation current with margin above the peak
• Acceptable DCR for the loss budget
• Board area, height limit and thermal environment

Design considerations

Thermal design deserves attention because the inductor sits among other heat sources. In a dense board the local ambient is elevated by neighbouring components, so the inductor is not starting from the board intake temperature. Its own loss adds to that, and heat leaves mainly through the terminations into the copper planes below.

The copper area connected to the pads is therefore part of the thermal design. A part that measures acceptably in free air can run considerably hotter when mounted on a small pad in a congested area.

Application examples

• Processor core and I/O rails on server and accelerator boards
• FPGA and ASIC supplies in networking and industrial equipment
• Memory and interface rails on dense digital boards
• Sensor and logic supplies in automotive modules
• Compact industrial controllers and embedded computing boards

Selection checklist

• Confirm inductance at the actual bias current, not the nominal value
• Check saturation current against worst-case transient load
• Balance DCR against the available thermal path
• Evaluate core loss at the real frequency and ripple waveform
• Verify the height and footprint fit the layout
• Measure temperature rise on the real board before design freeze

Custom development for POL designs

When a rail has a specific height limit or current level that no catalogue part satisfies, an application-specific inductor is often the more economical answer. Development covers core material and geometry, winding construction, termination design, and sample evaluation in the target layout, leading into production.

See the Co-Fired Molded Inductors guide for molded SMD series data, Molded Power Inductors for Automotive Electronics for the vehicle-side requirements, and Choosing Inductors for AI GPU VRM for the highest-density end of this application.

Frequently Asked Questions

What is a POL converter and why does it need a special inductor?

A point-of-load converter is placed close to the load it supplies and steps a distribution rail down to a low voltage at high current. Its inductor must deliver low DCR, stable inductance under bias and a low profile simultaneously, because it sits in the same dense area as the load.

Why are molded inductors common in POL designs?

Molded construction combines a shielded low-profile body, a distributed air gap for gradual inductance roll-off under bias, surface-mount termination and support for flat-wire windings that keep direct-current resistance low.

Does higher switching frequency allow a smaller POL inductor?

Yes, higher frequency reduces the inductance needed for a given ripple current, which permits a smaller component and improves transient response. The trade is increased core loss, so the core material has to be evaluated at the actual switching frequency and flux swing.

How do I choose between lower and higher inductance for a POL stage?

Lower inductance improves transient response but raises ripple current, core loss and the output capacitance needed. Higher inductance reduces ripple but slows response. The inductor and output capacitor network should be selected together against the load transient requirement.

What should be specified when sourcing a POL inductor?

Input and output voltage, steady-state and transient output current, switching frequency, target ripple and inductance, saturation current margin, acceptable DCR, and the available board area, height and thermal environment.