Inside an inductor

Unwind an inductor and you have a length of wire with a few milliohms of resistance. Wind it back up and it is a component. Nothing about the copper changed, so whatever the part does is being done by the shape.

Three things in that sequence are the whole page:

• The field is in the air, not in the wire. Through the middle of the coil, out of the end, around the outside, back in. Every turn's field lands on every other turn, which is why a coil does this and a straight piece of wire barely does.
• Once it has settled, the coil has stopped objecting. Steady current means a steady field, a steady field means nothing across the winding, and the part is back to being wire.
• The collapse goes the other way. That is not the same event played backwards; the voltage reverses, and it is as big as the interruption is fast.

Where the henries come from

L = µ₀ · µ_r · N² · A ÷ l

Four things you can change, and one of them is not like the others. N is the number of turns and it is squared, because each extra turn both adds to the field and gives the field something else to push against. A is the area the coil encloses and l is how long it is. µ_r is what is in the middle.

Twenty turns of air on a 10 mm former is about 1.6 µH. Drop a ferrite rod through the same twenty turns and it is hundreds of microhenries, for no extra wire and no extra space. That is the whole argument for cores.

What the core is doing

Iron and ferrite are full of tiny magnetic domains that line up with the field the coil makes, and their own field adds to it. The core is not storing the energy so much as concentrating it: it pulls the field lines in off the air and runs them through the middle, where they do some good.

Two things follow:

• The domains run out. There are only so many of them, and once they are all lined up the core has nothing more to give. That is saturation, it is the rating people misread, and it is the next page.
• Reversing them costs energy. Flip the field a million times a second and the core gets warm on its own account, before the copper has done anything. Solid iron is worse again, because a changing field drives currents in the metal itself. Ferrite is a ceramic and barely conducts, which is why the core in anything switching fast is ferrite and not iron.

Next: what is printed on the part, and the two current ratings it does not print.