Understanding the continuous current of an inductor requires looking past a single datasheet number. The maximum safe current is actually defined by two distinct physical limits: the thermal RMS current ($I_{rms}$), which dictates how much heat the copper windings can dissipate before the insulation melts, and the saturation current ($I_{sat}$), the threshold where the magnetic core loses permeability and inductance drops by a specified percentage (typically 20% or 30%). In practical DC-DC converter design, you must size your inductor so the average load current stays below $I_{rms}$, and the peak ripple current never exceeds $I_{sat}$.
The Core Rule: How Inductors Handle Current
An inductor resists changes in current according to the fundamental equation $V = L(di/dt)$. Think of it like a heavy flywheel in a mechanical drive train. Current is the rotational speed. You cannot change the speed instantly without either stripping the gears (generating a massive voltage spike) or stalling the engine (magnetic saturation).
When selecting a part, designers often fixate on the inductance value (e.g., 10µH) and ignore the current ratings. This is a fatal bench mistake. Datasheets specify two critical current thresholds:
- $I_{rms}$ (Thermal Limit): The DC current that causes the inductor's temperature to rise by a specified amount, usually $\Delta T = 40^\circ C$ above ambient. Exceed this, and the enamel on the magnet wire degrades, leading to shorted turns.
- $I_{sat}$ (Magnetic Limit): The peak current where the core material can no longer support additional magnetic flux. A 10µH inductor rated for 4A $I_{rms}$ might only have a 2.8A $I_{sat}$. If your peak circuit current hits 3.5A, the inductance collapses toward zero, ripple current spikes exponentially, and your switching MOSFET absorbs the excess energy and explodes.
Inductor Types and Current Handling Capabilities
Not all cores handle current the same way. The physical construction dictates the saturation curve, thermal dissipation, and electromagnetic interference (EMI) profile. Use this comparison matrix to select the right core for your topology.
| Type | Core Construction | Typical Tolerance | Tempco (ppm/°C) | Max Current Range | Typical Use |
|---|---|---|---|---|---|
| Ferrite Shielded | Ferrite drum + closed magnetic path | ±20% to ±30% | High (Non-linear) | 0.5A to 30A+ | High-current buck/boost converters, noise-sensitive RF boards |
| Ferrite Unshielded | Ferrite drum core, open magnetic path | ±20% to ±30% | High (Non-linear) | 0.1A to 15A | Low-cost, low-power DC-DC where EMI is not critical |
| Powdered Iron / Alloy | Distributed air gap in metal powder | ±15% to ±20% | Low (Linear soft roll-off) | 5A to 100A+ | High-ripple VRMs, automotive power stages, continuous conduction mode (CCM) |
| Ceramic / Air Core | No magnetic core, just wound wire | ±2% to ±5% | Very Low (Stable) | < 1A | High-frequency RF matching, GHz switching regulators |
Decoding Physical Markings and Safe Substitution
When you are digging through your parts bin, you need to read the EIA-style markings stamped on the top of the component. Most surface-mount power inductors use a three-digit code or an 'R' decimal indicator.
What the Markings Mean
- 100: 10 × $10^0$ = 10µH
- 471: 47 × $10^1$ = 470µH
- 2R2: The 'R' acts as a decimal point = 2.2µH
- R47: 0.47µH
Note: Some manufacturers print the exact value (e.g., "4.7") on larger shielded molded inductors, bypassing the code entirely.
How to Substitute Safely When the Exact Part is Missing
If you must swap core types, substituting a shielded inductor for an unshielded one is always safe (it reduces EMI). Substituting an unshielded part for a shielded specification will likely cause radiated emissions failures and can induce noise into nearby high-impedance analog traces.
Failure Modes: Visual Symptoms of Current Overload
When an inductor fails due to current limits, the visual evidence on the PCB tells you exactly which limit you violated.
Thermal Overload ($I_{load} > I_{rms}$)
The Symptom: The inductor is physically destroyed. You will see blackened epoxy, melted plastic casing, or charred PCB pads. If you scrape away the coating, the copper windings are fused together or burnt open. The switching IC and diode might actually survive because the inductor failed open-circuit, breaking the current path.
Magnetic Saturation ($I_{peak} > I_{sat}$)
The Symptom: The inductor looks perfectly pristine. No burn marks, no melted casing. However, the high-side switching MOSFET inside your DC-DC IC is shorted from drain to source, and the catch diode is blown.
The Physics: When the core saturates, inductance drops to near zero. The inductor essentially becomes a piece of straight wire connecting $V_{in}$ directly to the MOSFET. Current ramps up at thousands of amps per microsecond. The MOSFET hits its absolute maximum drain current ($I_{D(max)}$) and undergoes thermal runaway before the IC's overcurrent protection (OCP) can react. Saturation kills the silicon, not the coil.
Bench Walkthrough: A 12V-to-5V Buck Converter Meltdown
Let us look at a real-world scenario where ignoring the saturation current resulted in a dead board.
The Setup
We are designing a 5A synchronous buck converter stepping 12V down to 5V. The switching frequency ($f_{sw}$) is set to 300kHz. Based on a standard 30% ripple current target, the calculated ideal inductance is roughly 4.7µH. We pull a cheap, unshielded 4.7µH drum-core inductor from the bin. The datasheet says $I_{rms} = 7.2A$. Since our load is only 5A, we assume we have plenty of margin and solder it down.
The Numbers
Let us calculate the actual peak current the inductor will see.
Duty Cycle ($D$) = $V_{out} / V_{in}$ = $5 / 12 = 0.416$.
Ripple Current ($\Delta I_L$) = $\frac{(V_{in} - V_{out}) \times D}{L \times f_{sw}} = \frac{(12 - 5) \times 0.416}{4.7\mu H \times 300kHz} = 2.06A$.
Peak Inductor Current ($I_{peak}$) = $I_{out} + \frac{\Delta I_L}{2} = 5A + 1.03A = 6.03A$.
The Outcome
We power the board and slowly ramp the electronic load to 4A. Suddenly, the input supply trips into overcurrent protection. The buck IC is smoking. We desolder the inductor and test it with an LCR meter; it still reads exactly 4.7µH. The inductor is fine, but the $40 MOSFET driver IC is dead.
What Went Wrong
We looked at the $I_{rms}$ rating (7.2A) and ignored the $I_{sat}$ rating, which on that specific cheap drum core was only 5.2A. At 4A load, our peak current hit 5.03A. We were right on the edge of saturation. By 4.5A load, peak current exceeded 5.2A. The core saturated, inductance collapsed from 4.7µH to roughly 0.5µH, and the ripple current spiked to over 30A. The high-side FET shorted instantly. Always design for $I_{sat}$ using the peak current, not the average load current.
Selecting the Right Inductor for Your Next Build
To ensure your next power supply build survives the bench and the field, follow this decision framework:
- For high-current, continuous conduction mode (CCM) supplies: Choose powdered iron or metal alloy composite inductors (like the Vishay IHLP or Wurth WE-LQS series). They exhibit a soft saturation curve, meaning inductance rolls off gradually rather than falling off a cliff, giving your control loop time to react to transient spikes.
- For low-power, cost-sensitive consumer gadgets: Unshielded ferrite drum cores are acceptable, provided you maintain strict physical clearance between the inductor and any high-impedance feedback or analog traces to prevent magnetic coupling.
- For noise-sensitive RF or audio mixed-signal boards: Always pay the premium for fully shielded molded inductors (e.g., Coilcraft XEL or XAL series). The closed magnetic path keeps the AC flux contained inside the component.
For deeper mathematical modeling of inductor core losses and AC winding resistance (skin and proximity effects), refer to the design guides provided by Coilcraft Power Magnetics and the application notes available via the Wurth Elektronik Power Magnetics library. Additionally, All About Circuits offers excellent foundational calculus breakdowns for transient inductor behavior.
