The Same Inductance Can Mean Different Things
Inductance is not a single number. It varies with frequency, with DC bias current and with temperature. A datasheet that states an inductance value without stating the conditions under which it was measured is quoting a figure that may not describe the component in the application at all. Two suppliers can both be honest and still quote numbers that differ by tens of percent, because they measured under different conditions.
Test Frequency Changes the Result
Inductance is normally measured on an LCR meter at a specified frequency, commonly 100 kHz or 1 kHz depending on the component class. Core permeability and winding capacitance both vary with frequency, so the reading drifts. At frequencies approaching self resonance the apparent inductance rises before the component stops behaving inductively altogether. Comparing a value measured at 1 kHz with one measured at 100 kHz is not a like for like comparison.
- State the test frequency alongside every inductance value
- Compare figures only at the same frequency and signal level
- Bear in mind that self resonance raises apparent inductance before the part becomes capacitive
DC Bias Makes Power Inductors Complicated
For an inductor carrying DC current, the unbiased inductance is close to meaningless. As bias rises, permeability falls and inductance drops, gently for a powder core and abruptly for an ungapped ferrite core. A power inductor should be specified with an inductance versus bias curve, or at minimum with a stated inductance at a stated bias current, so the designer can confirm the value at the peak current the circuit will actually apply.
This is why the same nominal part can appear to meet a requirement in a catalogue and fail in a real converter. The catalogue value was measured with no bias.
DC Resistance, Quality Factor and Self Resonance
DC resistance is straightforward but must be referenced to a temperature, because copper resistance rises by roughly 0.4 percent per degree Celsius. A winding specified at 20 degrees Celsius will measure noticeably higher at 100 degrees. Quality factor expresses the ratio of stored to dissipated energy and is meaningful only at a stated frequency. Self resonant frequency defines the upper limit of inductive behaviour and should be comfortably above the highest frequency of interest in the application.
Fixture and Measurement Setup
Measurement fixtures contribute parasitic inductance and capacitance, and short leads matter more than they appear to. A four wire Kelvin connection eliminates lead resistance from the DC resistance reading but not from an inductance measurement, where fixture geometry and lead dress still influence the result. For very low inductance values the residual inductance of the fixture becomes a significant fraction of the reading, and the fixture must be compensated out.
How to Compare Two Specifications
Before concluding that one supplier offers a better component, confirm that the inductance figures were measured at the same frequency and signal level, that the difference between unbiased and biased inductance is accounted for, that DC resistance is quoted at the same temperature, and that quality factor and self resonance are specified under comparable conditions. Where a parameter is not stated, ask for it. A supplier that cannot supply conditions for its own datasheet values is not a supplier whose specification can be designed against.
Our specifications quote inductance with test frequency, DC resistance at a defined temperature, and inductance versus bias behaviour for power inductors, so that the values can be compared directly with the requirements of the circuit.