A saturated inductor is a magnetic component that has reached its maximum flux density ($B_{max}$). When the current through the coil exceeds the saturation current ($I_{sat}$), the core material can no longer support an increase in magnetic flux. The relative permeability ($\mu_r$) of the core plummets toward 1—the permeability of free air. Electrically, the inductance value collapses, and the component devolves into a low-resistance piece of wire. In switching power supplies, this inductance collapse causes the current slew rate ($di/dt$) to spike violently, often destroying the driving MOSFET in microseconds.
Datasheets typically define $I_{sat}$ as the DC bias current that causes a 20% or 30% drop in nominal inductance, depending on the manufacturer's convention. Understanding how different core materials handle this threshold, how to read physical markings, and how to diagnose failures is critical for any power electronics designer or bench technician.
Core Materials and Saturation Thresholds
Not all inductors saturate the same way. The physical construction and core chemistry dictate whether the inductor experiences a "hard" saturation (a sudden, cliff-like drop in inductance) or a "soft" saturation (a gradual, rolling decline). Choosing the right type for the job depends on your switching frequency, ripple current, and spatial constraints.
| Core Material | Construction & Air Gap | Saturation Profile | Tempco (Temp Stability) | Typical Use Case |
|---|---|---|---|---|
| Ferrite (MnZn) | Solid ceramic, discrete air gap required | Hard (sharp cliff) | Poor (Curie temp ~100°C-200°C) | High-frequency SMPS, RF chokes, low-cost buck converters |
| Powdered Iron | Insulated iron particles pressed together (distributed gap) | Soft (gradual roll-off) | Moderate | PFC chokes, low-frequency filtering, high-DC-bias applications |
| Sendust (Kool Mµ) | Fe-Si-Al alloy powder (distributed gap) | Soft (very gradual) | Good | Audio filters, high-DC bias inductors where audible whine must be eliminated |
| MPP (Molypermalloy) | Ni-Fe-Mo powder (distributed gap) | Soft (most linear) | Excellent | Military/aerospace, high-reliability resonant converters, precision filtering |
For high-frequency switch-mode power supplies (SMPS) operating above 500 kHz, MnZn ferrite is the standard due to its low core losses, but you must design with a strict margin below $I_{sat}$ because hard saturation is unforgiving. For applications with massive DC bias currents, like power factor correction (PFC) stages, powdered iron or Sendust cores are preferred. Their distributed air gaps prevent hard saturation, allowing the inductance to gracefully decline rather than collapse entirely. For a deeper look at core material magnetic properties, the Coilcraft core materials library provides excellent B-H curve comparisons.
Decoding Inductor Markings and Specs
Surface-mount power inductors, such as the popular Würth WE-PD or Coilcraft MSS1278 series, rarely have enough physical real estate for full part numbers. Instead, they rely on abbreviated stampings. Understanding these codes prevents costly bench mistakes.
Reading the Inductance Code
Most SMD inductors use a three-digit code based on picohenries (pH) or microhenries (µH), similar to ceramic capacitors but scaled differently:
- 470: 47 µH (47 followed by zero multipliers, or simply 47µH in standard notation).
- 101: 100 µH (10 followed by one zero multiplier).
- 4R7 or 4N7: 4.7 µH (The 'R' or 'N' acts as the decimal point).
- 222: 2.2 mH (22 followed by two zeros, yielding 2200 µH).
Distinguishing $I_{sat}$ from $I_{rms}$
When substituting or verifying a part, you must look at the datasheet to separate the magnetic limit from the thermal limit. Würth Elektronik's power magnetics guides heavily emphasize this distinction:
- $I_{sat}$ (Saturation Current): The magnetic limit. Exceeding this causes inductance collapse. It is largely independent of temperature.
- $I_{rms}$ (Thermal Current): The thermal limit. This is the DC current that raises the component's temperature by 40°C due to $I^2R$ copper losses and core losses. Exceeding this melts the solder joints or burns the winding insulation.
Bench Rule: Your peak inductor current ($I_{load} + \frac{1}{2}I_{ripple}$) must never exceed $I_{sat}$, and your continuous DC load current must never exceed $I_{rms}$.
A common mistake is selecting an inductor where $I_{rms}$ is higher than $I_{sat}$. In this scenario, the inductor will magnetically saturate and destroy your switching MOSFET long before the inductor itself gets hot enough to trigger a thermal shutdown. Always ensure $I_{sat} > I_{peak}$ and $I_{rms} > I_{dc}$.
Visual and Electrical Failure Modes of a Saturated Inductor
When an inductor is driven into deep saturation repeatedly, the collateral damage usually occurs to the surrounding circuitry rather than the inductor itself. However, the inductor will eventually show physical distress.
Electrical Symptoms
- Switch-Node Ringing and Spikes: As the core saturates, $L$ drops. Since $V = L(di/dt)$, a collapsing $L$ forces $di/dt$ to skyrocket to maintain the volt-second balance. This results in massive current spikes that trip the MOSFET's overcurrent protection (OCP) or punch through the silicon die.
- Audible Whining (Magnetostriction): Ferrite cores physically expand and contract with changing magnetic flux. If you hear a high-pitched squeal from a power supply under load, the inductor is likely operating near the edge of saturation, causing extreme magnetostrictive forces before hard clipping.
- Erratic Switching Frequency: In peak-current-mode controllers, premature saturation causes the current sense ramp to hit the controller's threshold prematurely, resulting in subharmonic oscillation or chaotic pulse-skipping.
Visual and Physical Symptoms
- Cracked Ferrite Cores: Hard saturation combined with high ripple currents generates localized thermal hotspots in the ferrite. The resulting thermal shock cracks the brittle ceramic core, often visible as a hairline fracture down the center leg of an EE or drum core.
- Melted or Discolored Potting Compound: If the inductor is encapsulated in epoxy (like molded shielded types), prolonged $I_{rms}$ overload or core-loss heating will turn the black/brown epoxy chalky, gray, or visibly melted.
- Desoldered Pads: Extreme thermal cycling from repeated saturation-recovery cycles can fatigue the solder joints, lifting the SMD pads off the PCB.
Safe Substitution When the Exact Part is Missing
Supply chain shortages frequently force engineers and repair technicians to substitute power inductors. You cannot simply swap parts based on the microhenry stamping. Follow this decision matrix to substitute safely:
- Match or Exceed $I_{sat}$ and $I_{rms}$: The replacement must have an $I_{sat}$ equal to or greater than the original. A 10µH inductor with a 5A $I_{sat}$ cannot be replaced by a 10µH inductor with a 3A $I_{sat}$, even if the physical footprint matches.
- Check the DCR (DC Resistance): If you substitute a part with a significantly higher DCR, you will increase $I^2R$ losses, dropping your converter's efficiency and potentially violating the $I_{rms}$ thermal limit.
- Maintain Shielding Topology: Never replace a shielded inductor (e.g., molded ferrite) with an unshielded drum-core inductor in noise-sensitive circuits. Unshielded parts leak fringing flux that will couple directly into nearby feedback traces or ADC lines, causing erratic regulation.
- Verify the Saturation Profile: If the original design relied on the soft saturation of a powdered iron core to handle transient load spikes gracefully, substituting a hard-saturating ferrite part will cause the power supply to trip OCP during those same transients.
Saturated Inductor FAQ
How do I measure if an inductor is saturated on the bench?
Standard handheld LCR meters apply a tiny AC test signal (usually < 1Vrms) with zero DC bias, meaning they will always read the nominal, unsaturated inductance. To measure saturation, you need an LCR meter with a DC bias current source accessory, or you must test the component in-circuit. In-circuit, place a current sense resistor in series with the inductor and view the voltage across it on an oscilloscope. A linear inductor will show a straight, sloped ramp for current. If the inductor is saturating, the current waveform will visibly "hook" or curve sharply upward at the peak of the switching cycle, indicating $L$ is dropping.
Can a saturated inductor recover once the current drops?
Yes. Magnetic saturation is a fully reversible physical phenomenon. Once the magnetizing force ($H$) drops below the saturation threshold, the magnetic domains in the core realign, and the inductance returns to its nominal value. The inductor itself is not permanently damaged by the magnetic saturation event. However, the current spike caused by the saturation event frequently destroys the driving semiconductor (MOSFET/IC) or blows the input fuse, which is what renders the circuit dead.
Why does my inductor get hot even if the current is below the saturation rating?
You are likely confusing $I_{sat}$ (the magnetic limit) with $I_{rms}$ (the thermal limit). An inductor can operate perfectly well below its saturation current but still overheat due to copper losses ($I^2R$ heating in the windings) and core losses (hysteresis and eddy currents in the magnetic material). Core losses scale non-linearly with switching frequency and AC ripple current ($\Delta I$). If your ripple current is too high, the core will overheat even if the peak current never touches $I_{sat}$.
Does an air-core inductor ever saturate?
No. Air has a relative permeability of exactly 1, and its B-H curve is perfectly linear all the way up to infinite field strengths. An air-core inductor will never experience magnetic saturation. However, because air lacks the flux-multiplying properties of ferromagnetic materials, air-core inductors have extremely low inductance values (usually in the nanohenry range). They are exclusively used in very high-frequency RF applications or extreme high-current pulse circuits where core losses and saturation must be entirely eliminated, regardless of the large physical size required to achieve the target inductance.
