Inductor circuits store energy in a magnetic field to resist changes in current. In switching power supplies, they smooth the output rail; in RF front-ends, they tune resonant frequencies; and in EMI filters, they choke high-frequency noise. But unlike resistors or ceramic capacitors, inductors are highly non-linear. Push them past their physical limits, and they don't just drift in value—they saturate, lose their inductance entirely, and turn your switching regulator into a dead short. Getting the right part requires looking past the nominal microhenry (µH) rating and understanding saturation current, RMS current, and core construction.
Decoding Inductor Markings and Specifications
Surface-mount device (SMD) inductors rarely have enough physical real estate for printed text. Instead, manufacturers rely on standardized three-digit or four-digit alphanumeric codes. If you are scavenging parts or verifying a reel, you need to read these markings fluently.
The standard EIA/JEDEC coding system works similarly to SMD resistors, but with a crucial difference in how the decimal is handled:
- Three-Digit Code (Whole Numbers): The first two digits are the significant figures, and the third digit is the multiplier (number of zeros). A marking of 100 means 10 × 10^0 = 10µH. A marking of 101 means 10 × 10^1 = 100µH. A marking of 472 means 47 × 10^2 = 4700µH (or 4.7mH).
- Code with 'R' (Decimal Values): The letter 'R' replaces the decimal point. A marking of R10 means 0.10µH. A marking of 4R7 means 4.7µH. A marking of R47 means 0.47µH.
- Tolerance Suffixes: You will often see a trailing letter indicating tolerance. J = ±5%, K = ±10%, and M = ±20%. Power inductors are almost always M (±20%), while RF inductors are typically J or K.
For a comprehensive breakdown of manufacturer-specific variations and physical sizing standards, the Coilcraft Inductor Guide remains the definitive bench reference for mapping physical footprints to electrical specs.
Inductor Types and Selection Matrix
Choosing the right inductor isn't just about the µH value; the core material dictates how the part behaves under load, temperature, and high frequencies. Here is how to match the construction to the job.
| Core Type | Construction & Material | Typical Tolerance | Tempco (ppm/°C) | Typical Use Case |
|---|---|---|---|---|
| Shielded Power | Molded ferrite or iron powder composite; magnetic flux contained within the body. | ±20% (M) | +100 to +300 | DC-DC buck/boost converters, high-density boards where EMI/crosstalk must be minimized. |
| Unshielded Power | Ferrite drum core with exposed wire winding; magnetic field extends into surrounding air. | ±20% (M) | +100 to +300 | Cost-sensitive power rails, LED drivers, where physical clearance from sensitive traces exists. |
| Ceramic Core | Non-magnetic ceramic substrate with thin-film or wirewound coils; no magnetic core material. | ±2% to ±5% (G, J) | +100 to +250 | RF matching networks, high-frequency filters (>100MHz), VCO tanks where high Q-factor is critical. |
| Ferrite Bead | Solid ferrite sleeve over a straight conductor; designed to dissipate high-freq energy as heat. | N/A (Impedance spec) | Varies widely | Power rail noise suppression, EMI filtering on I/O lines. Not for energy storage. |
Selection Rule of Thumb: Never use a ferrite bead in place of a power inductor in a switching regulator. Beads are designed to burn high-frequency energy as heat, not store it. Doing so will result in massive voltage droop and thermal failure. For a deeper theoretical look at how magnetic fields store energy across these different core materials, All About Circuits provides an excellent foundational primer.
Real-World Scenario: The Buck Converter Saturation Failure
Abstract datasheet parameters mean nothing until they fail on the bench. Let’s walk through a classic inductor circuit failure involving core saturation in a step-down (buck) converter.
The Setup
We are designing a 12V to 3.3V buck converter using a standard 500kHz switching regulator to power a 2A microcontroller load. Following the regulator’s datasheet formula, we calculate the required inductance:
- Vin: 12V
- Vout: 3.3V
- Iout: 2A
- Fsw: 500kHz
- Target L: 4.7µH
The Numbers
To select the inductor, we must calculate the peak current (I_peak), not just the average output current. First, we find the duty cycle (D) and the ripple current (ΔI_L):
- D = Vout / Vin = 3.3 / 12 = 0.275
- ΔI_L = (Vin - Vout) × D / (L × Fsw) = (12 - 3.3) × 0.275 / (4.7µH × 500,000) = 1.018A
- I_peak = Iout + (ΔI_L / 2) = 2A + 0.509A = 2.509A
The Outcome
We source a cheap 4.7µH unshielded drum-core inductor from a bulk bin. Its datasheet lists an Irms (RMS current rating) of 2.8A. Since 2.8A is greater than our 2A load, we assume it’s safe. We power up the board. Within 30 seconds, the switch node starts ringing violently, the output voltage sags to 2.1V, and the regulator IC becomes too hot to touch. The IC eventually trips its internal thermal shutdown.
What Went Wrong
We looked at Irms (thermal limit) but ignored Isat (saturation current). The cheap inductor we chose had an Irms of 2.8A, but an Isat of only 2.0A. When the inductor current hit 2.0A during the switching cycle, the magnetic core saturated. Its inductance instantly dropped from 4.7µH to near zero. With no inductance to limit the current slew rate (di/dt), the current spiked massively, overstressing the internal MOSFET of the regulator. Always ensure Isat is strictly greater than your calculated I_peak.
Failure Modes and Visual Diagnostics
When an inductor circuit misbehaves, the component itself rarely explodes like a tantalum capacitor. Instead, it fails in subtle ways that require specific diagnostic techniques.
- Open Circuit (Burnt Winding):
- Visual Symptom: Discoloration on the epoxy coating, a distinct burnt-resin smell, or visible blistering on the SMD pads.
- Bench Test: Multimeter reads OL (infinite resistance) across the pads. Caused by exceeding the Irms thermal limit, melting the internal copper wire.
- Shorted Turns (Insulation Breakdown):
- Visual Symptom: Looks perfectly normal to the naked eye. No burn marks.
- Bench Test: The DC resistance (DCR) drops significantly (e.g., from 0.05Ω to 0.01Ω), and an LCR meter shows inductance has dropped by 50% or more. Caused by high voltage spikes breaking down the thin enamel insulation between wire windings, creating an internal shorted loop that kills the magnetic field.
- Core Saturation (No Physical Damage):
- Visual Symptom: None. The part looks pristine.
- Bench Test: Requires an oscilloscope and a current probe. You will see the inductor current waveform change from a clean triangle wave to a triangle with a sharp, vertical "hockey stick" spike at the peak. Thermal imaging will show the inductor running unusually hot due to core losses.
- Mechanical Cracking:
- Visual Symptom: A hairline fracture running through the ferrite drum or shielding, often visible under 10x magnification. Common in large unshielded inductors subjected to board flexure or ultrasonic cleaning.
- Bench Test: Inductance may read correctly at rest but drops intermittently when the board is lightly flexed with a non-conductive probe.
Safe Substitution Rules When the Exact Part is Missing
Supply chain shortages frequently force engineers to substitute inductors. Swapping a resistor is trivial; swapping an inductor requires verifying five critical parameters to prevent circuit destruction.
Rule 1: Match the Inductance (Within Tolerance)
For power converters, a ±20% deviation in µH is usually acceptable, though it will alter your ripple current and loop stability. For RF matching or resonant tanks, you must match the exact value and tolerance (usually ±2% or ±5%).
Rule 2: Isat Must Exceed I_peak
This is non-negotiable. Check your circuit's peak current calculation (as demonstrated in the buck converter scenario). The substitute's saturation current (Isat) must be at least 20% higher than your calculated peak current to account for transient load steps and component aging.
Rule 3: Irms Must Exceed Maximum DC Load
Rule 4: Check the DCR (DC Resistance)
Lower DCR is generally better for efficiency. However, in some older voltage-mode buck controllers, a minimum DCR (or equivalent ESR) is required to generate the PWM ramp signal. If substituting a modern ultra-low DCR composite inductor into an older design, you may need to add a small series resistor or rely on the capacitor's ESR to maintain loop stability.
Rule 5: Shielding and Physical Footprint
Never substitute an unshielded inductor for a shielded one if the component is placed near sensitive analog traces, RF antennas, or Hall-effect sensors. The stray magnetic flux from an unshielded drum core will induce noise into adjacent loops. Conversely, substituting shielded for unshielded is electrically safe but may require pad modification, as shielded molded inductors often have different terminal geometries than wirewound drum cores.
