The Core Purpose of an Inductor: Opposing Change
If a capacitor's job is to oppose changes in voltage, an inductor's job is to violently oppose changes in current. The fundamental purpose of an inductor in any circuit is to store energy in a magnetic field and resist any attempt to alter the flow of electrons through it. This behavior is governed by Faraday's law of induction, expressed on the bench as:
Where V is the induced voltage, L is inductance in Henries, and di/dt is the rate of current change over time.
To understand why this matters, consider a practical scenario: you are switching off a MOSFET that is driving 5A through a 47µH inductor. If the MOSFET turns off in 50 nanoseconds, the rate of current change (di/dt) is 100,000,000 A/s. Plugging that into the formula yields a 4,700V voltage spike. This is exactly why flyback diodes are mandatory across relay coils and buck converter switches—without a path for that stored magnetic energy to dissipate, the inductor will arc across the switching element and destroy it.
Think of an inductor like a heavy mechanical flywheel connected to a motor. It takes significant effort (voltage) to get the flywheel spinning (current flowing), but once it is up to speed, it takes immense force to stop it abruptly. In power supplies, we use this 'flywheel' effect to smooth out chopped DC waveforms into clean, continuous current.
Inductor Types and Selection Criteria
Not all inductors are created equal. The core material dictates the saturation current, frequency response, and physical size. Choosing the wrong core for the job is the fastest way to fry a power stage. Below is a breakdown of the most common types you will encounter on a PCB or in a parts bin.
| Core Type | Construction / Material | Typical Tolerance | Saturation Behavior | Typical Use Case |
|---|---|---|---|---|
| Air Core | Copper wire wound on non-magnetic ceramic/plastic | ±2% to ±5% | Never saturates (linear) | RF tuning, high-frequency filters, crossover networks |
| Ferrite (MnZn) | Manganese-Zinc ferrite compound | ±10% to ±20% | Sharp, hard saturation | Low-to-medium frequency power supplies, EMI chokes |
| Ferrite (NiZn) | Nickel-Zinc ferrite compound | ±10% to ±20% | Hard saturation, high core loss at low freq | High-frequency (>1MHz) EMI suppression, RF chokes |
| Iron Powder | Insulated iron particles pressed into a core | ±10% to ±15% | Soft, gradual saturation (distributed air gap) | High DC bias applications, PFC chokes, buck converters |
| Shielded Drum | Ferrite drum enclosed in a magnetic shield sleeve | ±20% to ±30% | Moderate saturation, contains magnetic flux | Dense SMT power rails, noise-sensitive environments |
For a detailed look at core material losses and permeability curves, the Coilcraft Inductor Basics library provides excellent empirical data on how different compounds behave under DC bias.
Decoding Inductor Markings and Color Codes
Unlike resistors, inductor markings are notoriously inconsistent across manufacturers, but two dominant standards cover 90% of the parts on your bench.
SMD (Surface Mount) Numeric Codes
Most SMD power inductors use a three-digit code where the first two digits are the significant figures and the third digit is the multiplier (number of zeros), expressed in microhenries (µH).
- 100 = 10 × 10^0 = 10µH (Not 1µH, which is a common beginner trap)
- 101 = 10 × 10^1 = 100µH
- 472 = 47 × 10^2 = 4,700µH (or 4.7mH)
For values under 10µH, manufacturers use an 'R' to represent the decimal point. 4R7 means 4.7µH, and R47 means 0.47µH.
Through-Hole Color Bands
Axial through-hole inductors often use the standard 4-band resistor color code, but the base unit is microhenries, not ohms. A brown-black-brown-silver inductor reads as 1-0-1 (100µH) with a 10% tolerance (silver). Gold indicates a 5% tolerance. Always verify with an LCR meter, as faded paint on vintage or surplus parts can easily make a red (2) look like an orange (3).
Bench Scenario: When the Wrong Inductor Melts a Buck Converter
Theory is clean; the bench is unforgiving. Here is a real-world failure that highlights why understanding the purpose of an inductor extends beyond just its microhenry value.
The Numbers: In a buck converter, the inductor current is not a flat 3A; it is a triangle wave riding on top of the DC load. We designed for a 30% ripple current (0.9A peak-to-peak). This means the peak current ($I_{pk}$) the inductor must handle is:
$I_{pk} = I_{out} + (\Delta I_L / 2) = 3A + 0.45A = 3.45A$
The Mistake: The builder grabbed a generic 4.7µH SMD shielded drum inductor from a bin. The part had an $I_{rms}$ (thermal current rating) of 4.0A, which looked safe for a 3A load. However, its $I_{sat}$ (saturation current) was only 2.8A.
The Outcome: When the circuit powered up and the load hit 3A, the peak current of 3.45A pushed the ferrite core deep into magnetic saturation. Once saturated, the core could no longer store magnetic energy. The inductance instantly collapsed from 4.7µH to less than 0.5µH. With the inductance gone, the ripple current spiked from 0.9A to over 15A. The TPS54331's internal low-side MOSFET exceeded its absolute maximum current limit, shorted out, and permanently bricked the IC.
What Went Wrong: The builder confused thermal limits ($I_{rms}$) with magnetic limits ($I_{sat}$). For power inductors, the saturation current must always exceed the peak ripple current, not just the nominal DC output. For more on calculating power stage limits, refer to the TI Non-Isolated DC/DC Switching Regulators design portal.
Failure Modes and Visual Diagnostics
Inductors are generally robust, but they do fail. When troubleshooting a dead power supply or a noisy RF stage, look for these specific failure modes:
- Open Winding (Catastrophic Thermal Failure)
Visual Symptom: Cracked, bubbled, or charred epoxy coating. A distinct smell of burnt varnish.
Measurement: DCR (DC Resistance) reads infinite (OL) on a multimeter. The part is dead. - Shorted Turns (Insulation Breakdown)
Visual Symptom: Often looks perfectly fine externally, or shows slight bulging/discoloration on one side of the core.
Measurement: DCR reads slightly lower than the datasheet spec, but an LCR meter will show inductance has dropped by 20% or more. This happens when the thin enamel insulation between wire windings melts, creating a shorted loop that cancels out the magnetic field. - Core Cracking (Mechanical Stress)
Visual Symptom: Hairline fractures on the ferrite core, often near the PCB mounting pads.
Measurement: Inductance drops significantly. Ferrite is essentially compressed ceramic; it is highly brittle. If a PCB flexes during installation or thermal cycling, the core can snap. This introduces an unintended 'air gap', drastically lowering the permeability and the inductance.
Safe Substitution Rules When the Exact Part is Missing
You are repairing a board or prototyping, and the exact BOM inductor is out of stock. Can you substitute it? Yes, but you must follow this strict hierarchy of parameters to ensure the circuit survives.
Step 1: Match the Inductance (Within 20%)
For power conversion (buck/boost), a 10µH part can usually be swapped for a 12µH or 8.2µH part. The control loop will compensate, though ripple current and transient response will shift slightly. For RF tuning or precision filters, you must match the value exactly (±2% or better).
Step 2: Verify $I_{sat}$ vs. Peak Current
The substitute's saturation current ($I_{sat}$) must be equal to or greater than the original part. If you are unsure of the circuit's peak current, choose an inductor with an $I_{sat}$ at least 1.5 times the maximum DC load current.
Step 3: Check $I_{rms}$ (Thermal Limit)
The RMS current rating dictates how hot the part gets due to copper losses ($I^2R$). The substitute's $I_{rms}$ must meet or exceed the original. If you use a part with a lower $I_{rms}$, it will overheat and the enamel will eventually melt, causing a shorted-turn failure.
Step 4: DCR (DC Resistance) Should Be Lower
DCR causes voltage drop and heat. A substitute with a lower DCR is always safe and will improve efficiency. A substitute with a significantly higher DCR will reduce the maximum output voltage and cause the inductor to run hotter.
Step 5: Shielding and Physical Footprint
Never swap an unshielded inductor into a design that originally called for a shielded one if the inductor sits near sensitive analog traces, Hall-effect sensors, or high-gain op-amps. The stray magnetic flux from an unshielded drum core will induce noise directly into adjacent copper loops. Conversely, swapping shielded for unshielded is electrically safe, provided you have the physical clearance (shielded parts are typically taller).
