In DC steady-state, the potential drop across an inductor is simply the load current multiplied by its DC resistance ($V = I \times R_{DCR}$). However, during transient switching events, the potential drop is governed by Faraday’s law of induction: $V = L(di/dt)$. If you only design your circuit for the steady-state ohmic drop, the transient voltage spike generated during switching will inevitably avalanche and destroy your semiconductors. Understanding both regimes is the difference between a robust power supply and a smoking bench prototype.
The Physics of the Spike: Why Potential Drop Matters
To accurately calculate the potential drop across an inductor, you must separate the circuit's operation into two distinct time domains: steady-state and transient.
1. Steady-State (DC): Once the magnetic field is fully established and current is constant ($di/dt = 0$), the inductor acts merely as a low-value resistor. The potential drop is entirely determined by the wire's DC resistance (DCR). For a 10A load flowing through an inductor with a 20mΩ DCR, the drop is a negligible 200mV. This drop manifests purely as heat ($I^2R$ losses).
2. Transient (Switching): When a MOSFET opens or closes, the current through the inductor changes rapidly. The inductor opposes this change by inducing a voltage. The formula is:
Example: A 10µH inductor interrupting 2A of current in 50 nanoseconds generates a potential drop of 400V ($10\mu H \times \frac{2A}{50ns}$).
This 400V spike adds directly to your bus voltage. If your switching FET is only rated for 60V, it will undergo catastrophic avalanche breakdown. As detailed in Analog Devices' magnetics design guidelines, managing this transient potential drop requires careful selection of the inductor's core material, physical layout to minimize parasitic trace inductance, and properly tuned snubber networks.
Real-World Scenario: When $di/dt$ Destroys a Switching FET
Theory is clean; the workbench is not. Here is a recent failure analysis from a custom 24V-to-5V buck converter design that highlights the dangers of ignoring transient potential drops.
- The Setup: A synchronous buck converter switching at 1MHz, using a 4.7µH shielded ferrite inductor (Coilcraft XEL series) and a 40V-rated MOSFET. The steady-state load was 3A.
- The Numbers: The inductor's DCR was 15mΩ. Steady-state potential drop: $3A \times 0.015\Omega = 45mV$. The designer assumed the switching node would cleanly transition between 24V and 0V.
- The Outcome: Upon applying full load, the high-side MOSFET instantly shorted, vaporizing the silicon and cracking the package. Replacing the FET and probing the switch node with an oscilloscope revealed a massive 65V ringing spike every time the FET turned off.
- What Went Wrong: The designer calculated the potential drop across the component, but ignored the loop. The physical distance between the inductor, the catch FET, and the input decoupling capacitors created 8nH of parasitic PCB trace inductance. During the 15ns turn-off transient, the combined inductance ($4.7\mu H + 8nH$) generated a voltage spike that exceeded the 40V FET rating. The fix required moving the input capacitors within 2mm of the FETs to shrink the high-$di/dt$ loop area, reducing the parasitic potential drop to a safe 12V spike.
Inductor Construction Types and Selection Criteria
Choosing the right inductor dictates how it handles both steady-state DCR drops and transient saturation limits. Below is a comparison of common power inductor constructions.
| Type | Core Material | Construction | Tolerance | Tempco | Typical Use |
|---|---|---|---|---|---|
| Shielded Ferrite | NiZn or MnZn Ferrite | Wirewound inside a magnetic shield can | ±20% | Low drift until Curie temp | High-density buck/boost converters, noise-sensitive RF paths |
| Unshielded Ferrite | Ferrite Drum Core | Wirewound on a bobbin, exposed magnetic path | ±20% | Low drift until Curie temp | Cost-sensitive DC-DC, non-critical filtering |
| Powdered Iron / Alloy | Distributed air-gap metal powder | Molded composite or toroidal | ±10% to ±20% | Highly stable, soft saturation | High-current VRMs, automotive load-dump filtering |
| Wirewound Ceramic | Non-magnetic (Air core) | Wire on a ceramic substrate | ±2% to ±5% | Near zero | RF matching networks, high-frequency snubbers |
Which Type for Which Job?
Use shielded ferrite when your switching node is near sensitive analog traces or high-gain op-amps; the magnetic containment prevents the alternating flux from inducing noise in neighboring components. Use powdered iron/alloy when your circuit experiences massive current surges (like a motor start or hot-plug event). Ferrite cores have a 'hard' saturation curve—once they hit their $I_{SAT}$ limit, inductance drops to near zero instantly, causing a catastrophic spike in potential drop and current. Powdered cores exhibit 'soft' saturation, gradually losing inductance and giving your control loop time to react.
Decoding the Silk Screen: How to Read Inductor Markings
Unlike resistors, inductor markings are not universally standardized across all manufacturers, but the EIA 3-digit code is the most common convention for SMD power inductors. Understanding what the markings mean is critical when scavenging parts or verifying BOMs.
- The 3-Digit Code: The first two digits are the significant figures, and the third digit is the multiplier (number of zeros) in microhenries (µH).
100= 10 × 10^0 = 10µH (This is the most common point of confusion; it is not 100µH).471= 47 × 10^1 = 470µH.152= 15 × 10^2 = 1500µH (or 1.5mH).
- The 'R' Decimal Indicator: For values under 10µH, 'R' replaces the decimal point.
R47= 0.47µH.4R7= 4.7µH.
- Tracking Codes: Many manufacturers (like Würth Elektronik or Murata) print a secondary 2-character alphanumeric code. This is typically a date code or factory lot traceability marker, not an electrical value. Consult the specific manufacturer's datasheet to confirm.
How to Substitute Safely When the Exact Part is Missing
When your BOM inductor is on a 26-week lead time, substitution requires strict adherence to three rules:
- Never compromise on $I_{SAT}$ (Saturation Current): You can substitute an inductor with a higher $I_{SAT}$ rating, but never lower. If the original part was rated for 4A saturation, a 3A substitute will saturate during peak load, collapsing the inductance and spiking the potential drop across the switching FET.
- Shielded can replace Unshielded, but not vice-versa: Dropping a shielded part into an unshielded footprint is electrically safe and improves EMI. Putting an unshielded part into a design that relied on magnetic shielding will likely cause erratic behavior in nearby Hall-effect sensors or RF antennas.
- Watch the SRF (Self-Resonant Frequency): Higher inductance values in the same physical footprint require thinner wire and more turns, which increases parasitic capacitance and lowers the SRF. If the SRF drops below your switching frequency, the inductor stops acting like an inductor and becomes a capacitor.
Failure Modes and Visual Autopsy on the Bench
Inductors rarely fail silently. When they do, the physical evidence on the PCB tells a specific story about the electrical abuse they endured. According to Würth Elektronik's RED EXPERT design guidelines, thermal and magnetic stresses leave distinct signatures.
1. Thermal Runaway (Visual: Discolored Epoxy and Melted Solder)
The Symptom: The black or gray epoxy coating on the inductor turns brown or chalky white. The solder fillets on the pads look dull, grainy, or have reflowed into a ball.
The Cause: You exceeded the $I_{RMS}$ (heating current) rating. The $I^2R$ losses in the copper windings generated more heat than the component's thermal mass could dissipate. This is a steady-state failure, driven purely by DCR potential drop.
2. Core Saturation (Visual: No Damage to Inductor, Blown FET)
The Symptom: The inductor looks pristine. However, the switching MOSFET next to it is cracked, or the PCB trace leading to the FET is vaporized.
The Cause: The peak current exceeded $I_{SAT}$. The core saturated, inductance plummeted to near-zero, and the inductor acted as a dead short to the DC bus. The potential drop across the inductor vanished, transferring the entire bus voltage directly across the FET's $R_{DS(on)}$, causing instantaneous thermal destruction.
3. Mechanical Fracture (Visual: Hairline Cracks in the Ferrite Drum)
The Symptom: Under a 10x loupe, you can see a microscopic crack running through the ferrite core, or the winding wire is severed at the termination pad.
The Cause: Ferrite is essentially compressed ceramic dust. It is highly brittle. This failure usually occurs during automated ultrasonic PCB cleaning, board depaneling (v-score snapping), or if the PCB flexes during installation into a tight enclosure. The crack introduces a massive, unintended air gap, drastically altering the inductance value and the potential drop characteristics.
Measuring the Drop: Oscilloscope Probe Pitfalls
If you attempt to measure the transient potential drop across an inductor using a standard oscilloscope probe with a 6-inch alligator ground lead, your measurements will be entirely fictitious.
The large loop area formed by the probe tip and the long ground wire acts as an antenna, picking up the high-$di/dt$ magnetic flux radiating from the inductor. This induced noise manifests as massive high-frequency ringing on your scope screen, often showing 'spikes' of 20V or more that do not actually exist across the component terminals.
The Fix: To accurately measure the true potential drop across an inductor during a switching transient, you must use a tip-and-barrel probe or a ground spring attachment. Solder a small bare wire loop directly across the inductor pads, and probe the loop using the shortest possible ground return path. This minimizes the pickup loop area, revealing the actual voltage drop and allowing you to accurately tune your snubber components.
