An inductor magnetico (the international and legacy search term for a magnetic core inductor) relies on a ferromagnetic core—typically ferrite, powdered iron, or an alloy like Sendust—to multiply the inductance per turn of wire. Without a magnetic core, you would need hundreds of turns of wire to achieve 10µH; with a high-permeability core, you can achieve the same inductance in a compact 5x5mm SMD footprint. The direct answer for selection is straightforward: use Nickel-Zinc (NiZn) ferrite for high-frequency RF filtering, Manganese-Zinc (MnZn) ferrite for general switching power supplies, and powdered iron or Sendust for high-DC-bias applications where core saturation is a primary risk.
Understanding the physics, physical markings, and failure modes of these components is what separates a successful bench prototype from a field-deployable power supply. Below is a practical guide to selecting, reading, substituting, and troubleshooting magnetic inductors.
Magnetic Core Materials: Which Type for Which Job?
The core material dictates the inductor's behavior under DC bias, its operating frequency range, and its thermal stability. The permeability ($\mu$) of the core concentrates the magnetic flux, but every material has a saturation point where permeability drops sharply. Choosing the wrong material for your circuit topology will result in excessive heat, switching noise, or catastrophic FET failure.
| Core Material | Construction Style | Typical Tolerance | Tempco (ppm/°C) | Typical Use Case |
|---|---|---|---|---|
| MnZn Ferrite | Shielded / Unshielded drum | ±20% (M) | +100 to +300 | General purpose DC-DC buck/boost converters (100kHz - 2MHz) |
| NiZn Ferrite | Multilayer ceramic / Toroidal | ±5% (J) to ±10% (K) | -500 to -1000 | High-frequency RF chokes, EMI filtering (>10MHz) |
| Powdered Iron | Toroidal / Molded radial | ±10% (K) to ±15% | +50 to +200 | High DC bias power filtering, PFC circuits, audio crossovers |
| Sendust (KoolMµ) | Toroidal / Shielded SMD | ±10% (K) | -20 to +60 | High-current switching regulators, solar MPPT inductors, low-loss applications |
Selection Framework: If your circuit operates above 5MHz, MnZn ferrite will suffer from excessive core losses (eddy currents and hysteresis); switch to NiZn. If your DC-DC converter pushes high continuous current (e.g., >5A) and you cannot afford the inductance drop-off that occurs when ferrite approaches saturation, specify a distributed air-gap material like Powdered Iron or Sendust. For authoritative design parameters on core losses, refer to the Coilcraft Inductor Basics guide or manufacturer-specific core loss curves.
Decoding Inductor Markings and SMD Codes
Unlike resistors, which use a straightforward color band or 3-digit EIA code, inductor markings can be notoriously inconsistent across manufacturers. However, most SMD power inductors follow a modified 3-digit or alphanumeric system indicating the nominal inductance in microhenries (µH).
The Standard 3-Digit and Alphanumeric System
- Two digits + a multiplier: The first two digits are the significant figures, and the third digit is the multiplier (number of zeros).
Example:100= 10 × 10^0 = 10µH.101= 10 × 10^1 = 100µH.472= 47 × 10^2 = 4700µH (4.7mH). - The 'R' decimal indicator: For values under 10µH, the letter 'R' replaces the decimal point.
Example:4R7= 4.7µH.R22= 0.22µH. - Tolerance suffixes: You will often see a trailing letter indicating tolerance.
M= ±20%,K= ±10%,J= ±5%,F= ±1%. Power inductors are almost universally ±20% (M), while RF chip inductors are typically ±5% (J) or tighter.
Physical and Manufacturer Cues
Not all inductors are marked. Molded power inductors (like the Würth Elektronik WE-LQS or Coilcraft MSS series) often rely on reel labeling rather than top-surface printing. In these cases, physical dimensions and shield construction are your clues. A shielded inductor will have a continuous magnetic epoxy or metal casing wrapping the sides, whereas an unshielded drum-core inductor will expose the copper windings on the sides. Always cross-reference the physical footprint (e.g., 6x6mm, 12x12mm) with a parametric search on distributor sites like Digi-Key or Mouser when the top marking is missing or illegible.
How to Substitute Safely When the Exact Part is Missing
Supply chain shortages frequently force engineers to substitute magnetics. Swapping an inductor is not as simple as matching the microhenry value. A safe substitution requires matching or exceeding four critical parameters: Inductance ($L$), Saturation Current ($I_{sat}$), RMS Current ($I_{rms}$), and DC Resistance (DCR).
Worked Substitution Example
Suppose you are repairing a 12V-to-5V buck converter that requires a 4.7µH inductor. The original BOM calls for a specific TDK part rated for $I_{sat}$ = 4.5A and $I_{rms}$ = 3.2A. You only have two 4.7µH alternatives on your bench:
- Part A: 4.7µH, $I_{sat}$ = 3.0A, $I_{rms}$ = 4.0A, DCR = 15mΩ
- Part B: 4.7µH, $I_{sat}$ = 5.2A, $I_{rms}$ = 2.8A, DCR = 35mΩ
The Decision: You must calculate the peak inductor current ($I_{peak}$) of your circuit. In a standard buck converter, $I_{peak} = I_{out} + (\Delta I_L / 2)$. If your load draws 2.5A and the ripple current is 1A, your $I_{peak}$ is 3.0A.
Part A has an $I_{sat}$ of 3.0A, which leaves zero margin for transient load spikes; the core will saturate during a step-load. Part B has an $I_{sat}$ of 5.2A (safe from saturation), but its $I_{rms}$ of 2.8A is slightly below the 2.5A continuous load plus ripple heating. Part B is the safer choice, provided you verify the thermal rise ($I^2R$ losses = $2.5^2 \times 0.035 = 0.21W$) does not exceed the PCB's thermal dissipation capacity. For deeper thermal modeling, the All About Circuits magnetics guide provides excellent formulas for calculating core and copper losses.
Failure Modes and Visual Diagnostics
Magnetic inductors are generally robust, but they are not immune to physical and thermal failure. When troubleshooting a dead power supply, inspect the inductor for these specific symptoms:
- Core Saturation (Invisible but Deadly): There is no visual symptom on the inductor itself. However, the downstream switching MOSFET or diode will be shorted or cracked. Diagnosis requires an oscilloscope: look for sharp, massive current spikes at the switch node during the 'on' time of the PWM cycle, indicating the inductor has lost its impedance.
- Magnetostrictive Cracking: Ferrite materials physically expand and contract as magnetic domains align (magnetostriction). In high-power, high-frequency applications, this mechanical stress can cause hairline fractures in the core. Visual symptom: A visible crack running through the dark ferrite material, often accompanied by an audible high-pitched whine before failure. A cracked core alters the effective air gap, drastically reducing inductance.
- Thermal Degradation and Delamination: If the $I_{rms}$ rating is exceeded, the copper windings overheat. Visual symptom: The epoxy coating or plastic overmold turns yellow/brown, smells of burning phenolic resin, or physically bubbles. In severe cases, the solder joints at the SMD pads will reflow and crack due to CTE (Coefficient of Thermal Expansion) mismatch.
- Winding Short Circuit: High voltage transients (like an automotive load dump) can breach the thin enamel insulation on the copper wire. Visual symptom: Often none externally. Measured with a multimeter, the DCR will read unusually low (e.g., 0.1Ω instead of 2.5Ω), and an LCR meter will show a massive drop in inductance and a near-zero Quality factor (Q).
Frequently Asked Questions
Why does my inductor magnetico whistle or sing under load?
This phenomenon, known as 'coil whine,' is caused by magnetostriction in the core and Lorentz forces acting on the windings. When a switching regulator enters burst mode, skip-cycle mode, or operates at an audible frequency (typically between 1kHz and 15kHz) under light loads, the magnetic field pulses at a rate the human ear can detect. The physical vibration of the core halves and the loose copper windings generates the acoustic noise. To fix this, ensure your converter's switching frequency is well above 20kHz, use a fully shielded inductor with tight epoxy potting, or apply a dab of RTV silicone or conformal coating to physically dampen the windings.
Can I substitute a shielded inductor with an unshielded one in a pinch?
Electrically, yes, provided the $I_{sat}$ and $I_{rms}$ ratings match. Electromagnetically, it is highly risky. Unshielded drum-core inductors radiate significant magnetic flux into the surrounding space. If placed near a sensitive analog-to-digital converter (ADC), a Hall-effect sensor, or an RF antenna trace, this leaking flux will induce noise and degrade your signal-to-noise ratio. Only use unshielded inductors if the PCB layout guarantees adequate physical clearance from noise-sensitive nodes, or if the circuit is entirely digital and noise-tolerant.
How do I test for core saturation on the bench without destroying my circuit?
Do not test saturation by simply increasing the load on your live PCB until it fails. Instead, use an LCR meter equipped with a DC bias current source (like the Keysight E4980A with a DC bias fixture). Sweep the DC current from 0A up to the rated $I_{sat}$ while monitoring the inductance. The saturation current is formally defined as the DC bias point where the inductance drops by 20% (or sometimes 30%, depending on the manufacturer's datasheet definition) from its zero-bias value. If you lack a bias fixture, you can build a simple curve tracer using a power supply, a high-power sense resistor, and an oscilloscope to measure the V/L di/dt slope, but a dedicated LCR meter is the safest and most accurate method.
