Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 62473: Magnetic Components — Measuring Methods of Inductance and DC Resistance
IEC 62473, published in its latest edition in 2023, defines standardized procedures for measuring the two fundamental parameters of magnetic components: inductance (L) and DC resistance (DCR). Whether you are designing power inductors for a buck converter, common-mode chokes for EMI filtering, or flyback transformers for isolated supplies, the measurement methodology prescribed by this standard directly affects the accuracy, repeatability, and comparability of your characterization data. This article unpacks the technical requirements, instrumentation choices, and practical considerations that every power electronics engineer must understand.
Why It Matters: Without a standardized measurement protocol, two laboratories measuring the same inductor can report inductance values differing by 10–15% due to differences in test frequency, signal level, and fixture parasitics. IEC 62473 eliminates this ambiguity by specifying the exact test conditions.
1. Inductance Measurement: Principles and Test Conditions
Inductance measurement under IEC 62473 is based on the impedance analysis method at a defined test frequency. The standard emphasizes that inductance is not a fixed property of a component; it varies with frequency, DC bias current, AC signal amplitude, and temperature. Therefore, the standard mandates that the measurement report must document all of these conditions.
1.1 Test Frequency Selection
The standard defines preferred frequencies depending on the component type and intended application: 100 Hz, 120 Hz, 1 kHz, 10 kHz, 100 kHz, and 1 MHz. For power inductors used in switching converters (100 kHz–2 MHz), the recommended test frequency is typically 100 kHz. For line-frequency transformers (50/60 Hz), 100 Hz or 120 Hz is specified.
Measurement Pitfall: Testing a ferrite-core power inductor at 1 kHz (as some generic LCR meters default to) will yield an inductance reading that can be 2–3× higher than the effective inductance at the 500 kHz operating frequency, because the core permeability drops significantly above a few hundred kilohertz. Always match the test frequency to the intended operating frequency range.
1.2 AC Signal Level
The AC test signal amplitude must be specified and controlled. For most ferrite components, the standard recommends a signal level low enough (typically 0.1–1 Vrms) to avoid driving the core into saturation while still providing adequate signal-to-noise ratio. For powder-iron and sendust cores, which exhibit a “soft” saturation characteristic, the test signal level can significantly shift the measured inductance value.
| Core Material | Recommended Test Signal | Frequency | Typical L Accuracy |
|---|---|---|---|
| MnZn Ferrite (power) | 0.5 Vrms | 10–100 kHz | ±3% |
| NiZn Ferrite (EMI) | 1.0 Vrms | 100 kHz–1 MHz | ±5% |
| Iron Powder (toroid) | 0.1 Vrms | 1–10 kHz | ±8% |
| Sendust (high-flux) | 0.25 Vrms | 10–100 kHz | ±5% |
| Amorphous/Nanocrystalline | 0.2 Vrms | 10–100 kHz | ±4% |
2. DC Resistance Measurement: Four-Wire Kelvin Method
For DCR measurement, IEC 62473 mandates the four-wire (Kelvin) sensing technique to eliminate lead and contact resistance errors. This is critical because the DCR of power inductors is often in the milliohm range (1–100 mΩ), where even a 10 mΩ lead resistance can introduce a 50% or greater error if a two-wire method is used.
2.1 Measurement Current
The test current for DCR measurement must be high enough to produce a measurable voltage drop (typically >1 mV) across the component but low enough to avoid self-heating that would shift the resistance value. Copper has a temperature coefficient of approximately +0.393%/°C, so a 10°C rise from inadequate heat sinking during measurement produces a 3.9% DCR error.
2.2 Thermal EMF Compensation
The standard recommends offset compensation techniques to cancel thermoelectric voltages generated at the junctions between the test leads and the component terminals. A common approach is to use a pulsed DC measurement or a current-reversal method where the average of forward and reverse measurements eliminates the thermal EMF offset.
Engineering Best Practice: When measuring DCR below 10 mΩ, use a dedicated micro-ohmmeter with at least 1 A test current capability. For production testing, implement a relative humidity-controlled environment because copper oxidation can increase contact resistance by 2–5% over a 6-month period. Apply a thin layer of gold or tin plating on test pads for long-term repeatability.
3. Fixture Design and Calibration Requirements
The measurement fixture is often the largest source of error in both L and DCR measurements. IEC 62473 specifies requirements for fixture compensation, including open-circuit and short-circuit calibration (for L measurements) and offset nulling (for DCR measurements). For SMD components, the fixture must provide a defined contact force to ensure consistent contact resistance.
| Fixture Type | Best For | Residual Inductance | Contact Resistance |
|---|---|---|---|
| Kelvin clips | Through-hole, large SMD | 10–50 nH | <5 mΩ |
| Test tweezers | Small SMD (0402–1206) | 1–5 nH | <20 mΩ |
| Custom test jig | Production environment | <1 nH | <2 mΩ |
| Probe station | Wafer-level / bare die | <0.5 nH | <1 mΩ |
Critical Error Source: Open/short calibration must be performed at the same plane as the DUT connection. If the calibration reference plane is at the instrument terminals but the DUT is connected via a 10 cm cable, the cable’s residual inductance (~10 nH/cm for a twisted pair) will add 50–100 nH to the measurement, which can be larger than the component’s own inductance for small-value SMD inductors.
4. Frequently Asked Questions
Q1: Can I use a multimeter to measure DCR instead of a Kelvin setup?
A two-wire multimeter can be used for coarse verification (e.g., checking for opens/shorts), but it is not acceptable for formal compliance to IEC 62473. The lead and contact resistance error in a typical handheld multimeter is 100–500 mΩ, which is larger than many power inductors’ DCR specification. Always use four-wire Kelvin sensing for measurements below 1 Ω.
Q3: What is the tolerance requirement for the LCR meter itself?
IEC 62473 recommends that the measurement instrument’s basic accuracy be at least 4× better than the required measurement uncertainty. For a component with a ±10% inductance tolerance, the instrument should have a basic accuracy of ±2.5% or better. Annual calibration with traceability to SI standards is mandatory.
Q4: Does the standard cover saturation current (Isat) measurement?
No, IEC 62473 specifically covers L and DCR only. Saturation current characterization falls under different standards or application-specific specifications. However, the inductance measurement method in this standard can be extended — when combined with a DC bias current source — to generate the L vs. IDC curve that defines the saturation behavior.
Q5: How does temperature affect the measurement?
Ferrite permeability changes by approximately 0.1–0.3%/°C near room temperature, and the Curie point (typically 200–250°C for MnZn ferrites) marks a catastrophic drop. IEC 62473 recommends measuring at 25 ± 5°C unless otherwise specified. For automotive-grade components (AEC-Q200), measurements at 85°C and 125°C are typically required, and the standard’s fixture design should accommodate a thermal chamber.
