Those produce sinusoidal(ish) voltage waveforms; a spark-ignition system
requires an accurately-timed pulse. In the case of a magneto, this is
generated by shorting an auxiliary low-voltage winding to 'slug' the
change of magnetic flux in the iron core while the magnets move to
another position. When the short is removed, the flux flips over to the
new value and generates a fast pulse in the high-voltage winding.
Because the induced voltage is proportional to the rate of change of
flux (as well as the number of turns), a much higher voltage is produced
than if the flux change were allowed to follow the relatively-slowly
changing field due to movement of the magnets.
Some early magnetos incorporated a mechanical mechanism which had a
spring coupling between the armature and the shaft. At cranking speed,
a pawl held the armature stationary while the engine rotated and the
spring was tensioned, then suddenly released it to give the armature a
flip. There were still make-and-break points to give the rapid flux
change, but the faster mechanical movement at (hand) cranking speed made
for more reliable starting. The armature reverted to continuous
rotation once the engine reached idling speed.
Some very slow speed engines (portable agricultural engines) used a
system whereby the magneto shaft was spring loaded to a central position
and not directly coupled to the engine rotation. A projection on one of
the valve rods caught in a lever on the magneto shaft and pulled it away
from the central position against the tension of the spring. At a point
corresponding to top-dead-centre, the projection disengaged and allowed
the magneto shaft to flip back under spring tension, thereby generating
a high voltage pulse. The magneto armature performed several cycles of
oscillation before coming to rest, ready to be wound up again.
As far as I know, that was the only system that did not use points or a
transistor to give the required sharp change of magnetism.
> All these devices used a coil of wire and a magnet to generate high
> voltages. The secret with ignition systems is where the magnet and
> coil is in relation to the position of the engine piston.
If it was that simple, why did nearly every spark-ignition engine
manufacturer use points, condenser and a double winding on the coil or
magneto throughout the 70 year period before transistors were available?