Diagram of an induction coil with hammer-action interrupter.
The induction coil was invented in the late nineteenth century as a source of high voltage for laboratory experiments. The first coils of the design shown here were developed by the instrument maker Rhumkorff. He made use of a lever commutator mounted on the coil base to enable the direction of the primary current to be reversed and switched on or off. The induction coil is basically a step-up transformer, but its efficiency is low due to its cylindrical design. Much of the primary flux is lost, however the cylinder shape enables the insulation to be better formed, and this is necessary for the larger coils that can produce sparks several centimetres in length. The induction coil is a pulse transformer, and operates by switching on and off rapidly a large DC current in the primary coil.
The resulting induced emf in the secondary coil is proportional to the rate of change of flux through the secondary windings, which is in turn proportional to the rate of change of primary current.
A large secondary voltage pulse is generated on the switch off cycle of the primary, and a smaller pulse is produced on the switch on. The switching can be done by a variety of contact breakers, and over the years the contact breaker itself has become a separate component from the actual induction coil, allowing a given coil to be used with a variety of different contact breakers. The most common type of breaker is that often seen mounted to the coil itself, and uses the magnetic attraction of the iron core to attract a sprung loaded hammer. Its operation is explained with reference to the following diagram.
When the current is switched on, the primary is energised and the iron core becomes strongly magnetic. The hammer, which is also iron, is attracts towards the core but in doing so, breaks the primary circuit by separating the platinum contacts. Since no current flows in the primary coil any more, the iron is no longer magnetic and the hammer springs back and remakes the primary circuit.
So the cycle continues. The capacitor connected across the platinum contacts improves the efficiency of the coil. When the contacts separate, the collapsing magnetic field in the primary induces a large back emf in that coil, and the only path for the back-current is across the air gap between the separating hammer contacts. The resulting arc pits these contacts over time. However, with the capacitor there, instead of causing an arc, the back emf transient momentarily charges the capacitor which then discharges through the primary. Since the polarity of this pulse is opposite to that of the original supply, it is as if the supply current had not only been switched from a constant value to zero, but actually traversed down below zero and to some value of the opposite polarity. Thus the effective change of current in the switch-off is much larger than it would be if the capacitor was not present.
The factor improvement in performance is significant. An induction coil without a capacitor will produce small secondary parks and its contacts will wear out within minutes of use. A coil with a capacitor fitted will produce far longer secondary sparks and its contacts will last for hours.
We have built an induction coil that produces secondary sparks 10cm long with 24V applied to the primary circuit. The primary coil is wound around a turfnol former around 25mm diameter. The primary is wound from enamelled copper wire 1.2mm thick (grade II high temperature type) and contains around 200 turns of wire. The secondary is wound from 0.04mm thick enamelled copper wire and contains 400000 turns of wire (around 60km of wire). The wire was wound in layers of 10000 turns and then a layer of insulation consisting of several wrap-rounds of polythene sheeting and insulating tape is wound on before the next 10000 turns. In total there are around 40 layers of insulation. The dimensions of the coil are actually quite small, and the wide appearance of the finished article is largely padding so that there is a sizeable distance from the core to the points where the physical secondary contacts are made. One end of the secondary (that nearest the core) is connected to one end of the primary and this tied to a common earth, as in the diagram above.
The core is about 300mm long, with the primary extending over that length, but the secondary coil begins about 6cm in from the ends to avoid the problem of sparking round the ends of the coil. There is another turfnol tube between the primary and secondary to give further insulation The capacitor is a car contact-breaker condenser, which is designed for the very same purpose in car ignition systems. Its polarity in the circuit should not matter as it is a non-polarised capacitor. The coil can be fed from around 100V AC at a few amps for small periods of time (a second or two in every minute). This produces a very fiery arc suitable for Jacob Ladder effects.