The Induction Coil

Wind a primary of a few dozen turns and a secondary of several hundred on the same iron core, and connect a sensitive meter across the secondary.

Switch the primary on and leave it on. Watch the meter.

Expect a flick at the moment of switching and then nothing, however long you wait and however large the current.

Write the conclusion plainly, because everything else follows from it: a magnetic field does not induce a voltage. A CHANGING magnetic field does. A steady current is, to the secondary, indistinguishable from no current.

Switch the primary on and off repeatedly and watch the secondary carefully at each event.

Expect the break to give a much bigger kick than the make.

The reason is asymmetry in time, not in the coil. Closing a switch builds the field at a rate the circuit's own inductance limits; opening one collapses it as fast as the contacts can separate, which is far faster.

This is why every induction coil is designed around the interruption and not the connection, and why the output is a series of sharp pulses rather than anything resembling the smooth supply that drives it.

Arrange a springy contact so that the primary current, flowing through the electromagnet, pulls the contact open — which cuts the current, which releases the contact, which closes it again.

Expect a buzz and a rapid train of output pulses.

You have built an oscillator out of nothing but a switch that turns itself off, and it is exactly the arrangement in an electric bell.

Note the elegance and the flaw together: it needs no separate timing device, and its rate is set by whatever the spring and the magnet happen to agree on, which is a poor way to control anything.

Run the coil and watch the interrupter contacts closely. Then connect a capacitor across them and compare.

Expect a visible arc and pitted contacts without it, a much sharper break and a bigger output with it.

The reason is worth stating carefully. An arc is the current refusing to stop — it keeps flowing through ionised air after the metal has parted, so the field collapses slowly and the output is feeble. The capacitor gives that current somewhere to go for the microseconds it needs, so the arc never strikes.

It is not a smoothing component here. It is there to make an event sharper, which is nearly the opposite of what capacitors usually do.

Measure the output spark gap you can sustain, then compute the turns ratio and the voltage it predicts.

Expect the real output to exceed the turns-ratio prediction, substantially.

Then slow the interrupter deliberately — soften the break with a poor contact — and watch the output fall even though the turns are unchanged.

The transformer rule is a statement about steady alternating current. An induction coil is not run that way, so its output is governed by the speed of one violent transient. The winding sets the scale; the interrupter sets the answer.

Treat the output as dangerous even from a small battery: it is the voltage, not the source, that jumps a gap.

The idea is Faraday's; the instrument is a chain of improvers. Nicholas Callan, a priest and physicist at Maynooth, built large coils from the 1830s. Heinrich Ruhmkorff, an instrument maker in Paris, refined the construction and insulation around 1851 to the point where the coil became a reliable laboratory product — which is why it is usually called a Ruhmkorff coil even though he was not first. Making a thing manufacturable is a real contribution, and it is the one most often mistaken for invention.

It was the standard high-voltage source for half a century. Every X-ray tube, every gas-discharge experiment, every spark-gap radio transmitter — including the one in the Marconi Wireless blueprint — ran on one. Röntgen discovered X-rays using a coil in 1895, so this device is upstream of a whole branch of medicine as well as of radio.

The interrupter was the weak point, and the history is a hunt for better ones. The self-buzzing contact wears out and limits the rate; mercury turbine interrupters, electrolytic Wehnelt interrupters and motor-driven breakers were all attempts at the same problem — break the current faster, more often, and without destroying yourself. That is the same requirement as a switching power supply's transistor today, and the modern answer is the same trick with no moving parts.

It is the ignition coil in a petrol engine. Unchanged in principle: low-voltage supply, two windings, an interrupter (contact breaker, later a transistor), a capacitor across it to stop arcing, and a spark at the output. A component invented for the laboratory bench ended up in several billion vehicles doing precisely the job it was invented to do.

Honest limits. It produces sharp unidirectional pulses, not a controllable waveform, so it is a poor supply for anything that wants steady high voltage. Output depends on the interrupter's behaviour, which drifts as contacts wear. Insulation is the real engineering difficulty — thousands of adjacent turns at thousands of volts want to arc between layers, which is why the secondary is wound in sections. And the sharp break radiates broadband interference in every direction, which is delightful if you are building a spark transmitter and intolerable for everybody else.