The Core Function: Why We Use Inductors in Modern Circuits
An inductor stores energy in a magnetic field when electrical current flows through it, fundamentally resisting any change in that current. The governing equation is V = L(di/dt). If you try to change the current instantly, the inductor generates a massive voltage spike to fight it. In a practical 5V-to-3.3V buck converter switching at 500kHz, a 10µH inductor smooths the chopped DC into a steady output current by absorbing energy during the switch's 'on' time and releasing it during the 'off' time.
Unlike resistors that burn energy as heat, or capacitors that store energy in an electric field, inductors deal strictly in magnetic flux. This makes them indispensable for power conversion, RF filtering, and choke applications. However, selecting the right part requires balancing inductance (L), DC resistance (DCR), saturation current (Isat), and RMS current (Irss). According to Coilcraft's magnetics design guides, confusing Isat (the point where the core magnetically saturates and inductance drops) with Irss (the thermal limit where the wire overheats) is the most common mistake in power supply design.
Inductor Types and Selection Criteria: Which Core for Which Job?
Not all inductors are created equal. The core material dictates how much energy the part can store before saturating, how it behaves at high frequencies, and its physical footprint. Here is how to map core types to specific circuit jobs.
| Core Type | Construction | Tolerance | Tempco (ppm/°C) | Typical Use Case |
|---|---|---|---|---|
| Air / Ceramic Core | Wire wound on non-magnetic ceramic or plastic former | ±2% to ±5% | Very Low (<50) | VHF/UHF RF tuning, high-Q filters, GHz matching networks |
| Ferrite (Shielded) | Molded ferrite powder completely enclosing the coil (e.g., Würth WE-LQS) | ±20% to ±30% | Moderate (100-300) | DC-DC buck/boost converters, point-of-load regulators |
| Metal Alloy Powder | Distributed gap metal powder core (e.g., TDK SPM series) | ±20% | Low to Moderate | High-current, fast-transient POL converters, automotive rails |
| Multilayer Ceramic | Low-Temperature Co-fired Ceramic (LTCC) printed spiral layers | ±5% to ±10% | Stable | Tiny 0402/0603 RF chokes, Bluetooth/WiFi impedance matching |
| Toroidal Iron Powder | Donut-shaped pressed iron powder with distributed air gaps | ±15% | High (varies by mix) | Power Factor Correction (PFC) chokes, high DC bias line filtering |
Decoding the Markings: How to Read Inductor Codes and Specs
When you are scavenging parts from a reel or trying to identify a component on a salvaged PCB, the printed codes can look like gibberish. Most SMD power inductors use a three-digit alphanumeric code based on microhenries (µH).
- The 'R' Decimal System: If the value is less than 10µH, an 'R' represents the decimal point.
4R7means 4.7µH.R10means 0.10µH (100nH). - The Three-Digit System: For values 10µH and above, the first two digits are the significant figures, and the third digit is the multiplier (number of zeros).
100means 10 followed by zero zeros = 10µH.101means 10 followed by one zero = 100µH.472means 47 followed by two zeros = 4700µH (4.7mH). - Nanohenry RF Codes: Small RF chip inductors (like the Murata LQG series) often mark nanohenries. A marking of
10Nor100(in a specific RF context) might mean 10nH. Always check the manufacturer datasheet for RF parts, as the coding overlaps with the microhenry standard.
Beyond the inductance value, you must read the spec sheet for the hidden parameters. A 10µH inductor with a DCR of 0.05Ω and an Isat of 5A will perform vastly differently under load than a 10µH inductor with a DCR of 0.5Ω and an Isat of 1A, even if their physical footprint is identical.
Bench Failures: Visual Symptoms and Root Causes of Inductor Death
Inductors are generally robust, but they fail catastrophically when pushed past their physical limits. Here is what failure looks like on the bench and under the oscilloscope probe.
1. Thermal Runaway (Exceeding Irss)
Visual Symptom: The epoxy encapsulation turns dark brown or black, cracks, or emits a distinct burning phenolic smell. In severe cases, the solder pads lift off the PCB due to prolonged heat transfer down the leads.
Root Cause: The RMS current exceeded the thermal rating of the copper wire. The I²R losses generated more heat than the component's surface area could dissipate. This often happens when a power supply is subjected to a continuous overload condition without adequate hiccup-mode protection.
2. Core Saturation (Exceeding Isat)
Visual Symptom: There is often no visual damage to the inductor itself. However, the downstream switching MOSFET or controller IC will be physically cracked or blown open. On a scope, the inductor current waveform changes from a clean triangle wave to a sharp, vertical spike at the end of the switching cycle.
Root Cause: When the core saturates, the magnetic permeability drops to near that of air. The inductance effectively collapses to zero. With no inductance to limit the rate of current rise (di/dt), the current spikes to hundreds of amps in nanoseconds, instantly destroying the switching FET.
3. Mechanical Fatigue and Micro-Cracking
Visual Symptom: A microscopic crack in the ferrite core material, visible only under 10x magnification, or a fractured solder fillet at the PCB pad.
Root Cause: Magnetostriction. Ferrite materials physically change shape slightly when magnetized. In high-power audio amplifiers or switching supplies operating in the audible frequency range (1kHz - 20kHz), this causes the component to 'sing' or whine. Over years of thermal cycling and mechanical vibration, this stress fractures brittle ferrite cores.
Substitution Rules: Swapping Parts When the Exact BOM Line is Obsolete
Supply chain shortages frequently force engineers and repair technicians to substitute inductors. You cannot just swap a 10µH part for any other 10µH part. Follow this strict hierarchy to substitute safely:
- Match Inductance (±20%): The replacement must be within the tolerance band of the original. A 20% drop in inductance will increase the ripple current in a buck converter, potentially tripping overcurrent protection or increasing output voltage ripple.
- Verify Isat Margin: The replacement's saturation current must be equal to or greater than the original. Calculate the peak current of your circuit (Iout + half the ripple current) and ensure the new Isat is at least 20% above that peak.
- Check DCR and Irss: A replacement with higher DC Resistance will run hotter and drop more voltage. Ensure the replacement's DCR is equal to or lower than the original, and its RMS current rating meets the continuous load requirement.
- Shielding Compatibility: Never replace a shielded inductor with an unshielded drum-core variant in a mixed-signal design. The unshielded part will radiate EMI, potentially failing FCC/CE compliance or disrupting nearby RF antennas.
- Footprint and Height: Verify the pad layout matches exactly. Forcing a part with a different pad pitch will result in weak solder joints that fail under thermal expansion. Check the Z-axis height to ensure the enclosure lid will still close.
For deep technical parameters on metal alloy versus ferrite saturation curves, refer to the TDK Electronics magnetics library, which provides excellent comparative B-H curve graphs for substitution analysis.
Frequently Asked Questions About Inductor Selection and Behavior
Why does my switching regulator inductor whine or squeal under light loads?
This is caused by magnetostriction and sub-harmonic oscillation. When a switching regulator enters 'pulse-skipping' or 'burst' mode at light loads to save power, the switching frequency drops into the human audible range (20Hz to 20kHz). The magnetic core physically expands and contracts at this frequency, acting as a tiny speaker. To fix this, select an inductor with a molded, fully shielded construction (which dampens acoustic noise) or use a regulator that enforces a fixed switching frequency above 20kHz even at light loads (often called 'forced PWM' mode).
Can I put two 10µH inductors in parallel to double the current rating?
Yes, but with strict caveats. Two 10µH inductors in parallel yield 5µH of total inductance, and they will theoretically share the current. However, if their DC resistances (DCR) are not perfectly matched, one will hog the current and overheat first. Furthermore, if they are placed physically close together, their magnetic fields will couple (mutual inductance), drastically altering the effective inductance and potentially causing localized core saturation. If you must parallel them, use parts with a positive temperature coefficient for DCR (like standard copper wire) so that as one heats up, its resistance increases, naturally balancing the current share.
What exactly happens to the circuit if I use an inductor with a lower saturation current than specified?
When the current reaches the Isat limit, the magnetic core can no longer store additional flux. The inductor stops acting like an inductor and behaves like a low-value resistor (essentially just the DCR of the wire). In a switching power supply, this removes the di/dt limiting effect. The current through the switching MOSFET ramps up almost vertically, instantly exceeding the FET's maximum drain current and avalanche energy ratings, resulting in a shorted, exploded, or melted semiconductor.
Does the physical orientation of an SMD inductor matter on the PCB layout?
For fully shielded molded inductors, orientation matters very little. However, for unshielded or semi-shielded drum-core inductors, orientation is critical. These parts have an open magnetic path (usually at the top or sides). You must orient the part so that the 'open' magnetic flux lines do not cross sensitive, high-impedance analog traces, feedback nodes, or RF antenna feeds. Always consult the manufacturer's datasheet for the specific magnetic flux leakage diagram of the part you are using.
