Inductance Converter
Same value in every unit
| Unit | Value |
|---|
Reading an inductance value without slipping a factor of 1000
Most inductance mistakes are not arithmetic mistakes. They are a µ read as an m, or a marking taken at face value. Inductance in ordinary electronics spans about ten orders of magnitude, from a bond wire of roughly a nanohenry to a mains choke of several henries, so the prefix carries nearly all the information in the number.
Inductance resists a change in current, the mirror of what capacitance does for voltage. The defining relationship is V = L × di/dt: one henry develops one volt when the current through it changes at one amp per second. Steady DC develops no voltage across an ideal inductance, which is why a coil that behaves perfectly on a bench supply can still ruin a switching edge.
Where each prefix actually turns up
- Picohenries: below the level of real components, used mainly in package, via and interconnect models.
- Nanohenries: PCB traces, package leads, bond wires, RF matching and tuning coils.
- Microhenries: buck and boost converter inductors, RF chokes, HF and VHF tuned circuits.
- Millihenries: audio and filter chokes, relay and solenoid windings, larger switch-mode output inductors.
- Henries: mains transformer windings, line-frequency chokes, large iron-cored coils.
Reactance anchors the number to a frequency: X = 2πfL. A 10 µH inductor is about 63 Ω at 1 MHz and about 0.63 Ω at 10 kHz. Same component, entirely different role in the circuit.
The nanohenries you did not design in
A short PCB trace carries stray inductance in the low nanohenries. A working rule of thumb is on the order of a nanohenry per millimetre for a narrow isolated trace; running it tightly over a ground plane brings that down, and the real figure depends on width, height above the plane and where the return current actually flows. That seems ignorable until a fast edge goes through it. At a di/dt of 1 A/ns, a single nanohenry develops a volt. That is the entire argument for short decoupling loops, unbroken ground planes and tight gate-drive returns: you are minimising an inductance that never appeared on the schematic.
Abhenry and stathenry
Both are CGS leftovers. The abhenry, from the electromagnetic system, is defined as exactly one nanohenry, so a figure in abhenries transfers across unchanged. The stathenry sits at the other extreme, 8.98755 × 10^11 H, which is c squared times 10^-9 with the speed of light in cm/s: an exact figure, because the electrostatic system ties its units to c. No real component comes near it. It is here so that older references and conversion tables can be read, not because anything is specified in it.
Share this tool with friends
Free to use, no sign-up, works on any phone.
Frequently Asked Questions
One thousand. 1 mH = 1000 µH, and 1 µH = 1000 nH, so every step between these named prefixes is a factor of 1000 rather than 10 or 100. That is why a misread prefix is never a small error. If a calculated LC corner frequency comes out about 32 times off, suspect a prefix slip: frequency depends on the square root of inductance, and the square root of 1000 is 31.6.
Yes, exactly. One abhenry is defined as 10^-9 H, which is one nanohenry with no rounding involved. The abhenry belongs to the CGS electromagnetic system and now survives mainly in older textbooks and physics tables. If a source quotes a value in abhenries, you can read the same number as nanohenries directly.
The stathenry is the CGS electrostatic unit of inductance, equal to 8.98755 × 10^11 H, or roughly 900 billion henries. That figure is exact rather than measured: it is c squared times 10^-9 with the speed of light expressed in cm/s, because the electrostatic system defines its units through c. Nothing you can physically build comes anywhere near that value, so it never appears on a datasheet. It is included here only so that old references can be converted.
On the usual three-digit code it means 10 followed by one zero, so 100, and for inductors the implied unit is normally microhenries: 100 µH. This catches people out because the same code on a ceramic capacitor implies picofarads instead. Small values often use R as a decimal point, so 4R7 reads as 4.7 µH. Marking conventions do vary between manufacturers, so check the datasheet before committing a value.
Because the voltage across an inductance depends on how fast the current changes, not on how much current flows. With V = L × di/dt, one nanohenry carrying an edge of 1 A/ns develops exactly one volt. Modern switching devices reach that range easily, so a couple of centimetres of return path can produce visible overshoot and ringing. It is a layout problem, not a component problem.
Use X = 2πfL, with L converted to henries and f in hertz. For example, 100 µH at 100 kHz gives 2π × 100000 × 0.0001, about 62.8 Ω. Convert the prefix first, before multiplying, since that is where most errors creep in. Remember that this is the ideal-inductor reactance: a real inductor also has a self-resonant frequency, above which its parasitic capacitance dominates and it stops behaving inductively.