A Sort-of Explanation of

High-tension Ignition

Since there is some confusion about high-tension ignition systems, I've made a series of oscilloscope photos to try to explain how it works. 

I will also explain why you don't get much of a spark at the plug and a big spark at the points when there is no condenser in the circuit.  It explains, through the back door, how a low tension ignition system works.

NOTE:  Click on the drawings for a clearer view.

Anyway, here's a drawing of a standard point-coil-condenser setup like you have on your older cars.

Here's the same thing, only in a regular schematic format.

I won't get too technical here but, the main component of the system is a high-tension induction coil.  It is shown in the top drawing about as it appears on the car.  In the bottom photo, it is shown as two coils of wire (the "primary" and "secondary") separated by some straight lines which represent what is called the core..  

A high-tension induction coil is made by first making the core, which is usually either iron wire or iron sheets.  Insulation is wrapped around the core and the primary winding (on the left in the above drawing) is wound on this insulation layer.  The primary is composed of a relatively small number of turns of relatively heavy wire.

Once the primary is wound over the core, another layer of insulation is applied.  Now, the secondary (the winding on the right in the above drawing) is wound.  The secondary is composed of a large number of turns of very fine wire.  One end of the secondary winding goes to the frame of the coil (the can) and the other end goes to the tower where the plug wire is inserted.

Before sealing the coil in the can, it is usually vacuum impregnated with insulating oil.

The way a high-tension induction coil works is to transform a lower voltage in the primary winding into a much higher voltage in the secondary.  Since the primary is made of a small number of turns of heavy wire, when it is connected to a battery, a lot of current is drawn and a lot of magnetic lines of force are produced (magnetism).  The core focuses the magnetism.  As the magnetism builds-up and decays, the changing magnetic lines of force cut across the many turns of the secondary, producing a short high voltage low current pulse that jumps across the gap in the spark plug.

Now, in itself, this setup will produce a small high voltage spark at the plug but the points will burn because, when the source of current is removed from the primary, the collapsing magnetic field will induce several hundred volts back into the primary.  This voltage rises extremely fast, faster than the points can open.  An arc is formed as the points open and the heat generated by this arc is what burns the points.

Placing a capacitor ("condenser") across the points does two things.  First, it makes the voltage build-up across the opening points much slower so there is a much smaller arc produced.  Smaller arc means less heat and less burning.  Second and very important is the fact that the condenser "tunes" the primary so the voltage builds up and decays in what is called a "damped oscillation" when the points open. 

When the points close, the condenser is shorted out and full voltage is impressed on the primary.  The current in the coil doesn't instantly build-up because, as the magnetic field expands, it generates what is called a "back EMF" in the winding which, as it expands, counteracts the current build-up.  Once the field is near maximum, the back EMF decreases because the field is no longer changing fast and the current builds-up.

Current build-up in the primary winding of the induction coil.

The above oscilloscope photo shows the current build-up after the points close.  Note that it takes less than 1/100th of a second for the current to go from zero to 3.5 Amperes.

Now, when the points open, there is a completely different action because the condenser is no longer being shorted-out by the points.

The time line runs from left to right.  The points open a little over one division from the left.

In the above scope picture, the top trace is the voltage across the primary of the coil (across the points) and the bottom trace is the voltage across the spark plug.  Note that, since I do not have a suitable high voltage probe for my scope, I simply hooked a heavily insulated wire over the plug wire to capacitively couple the voltage to the probe.  Necessarily, I got a very inaccurate (in other words, useless) trace of the high voltage.

The oscillation of the voltage across the primary is caused when the collapsing magnetic field generates a voltage.  The condenser resists the voltage build-up as it charges.  This slows down the collapse of the field.  When the field voltage decreases to that of the condenser, the condenser pumps it's charge back into the primary and so on.  The voltage decays in subsequent cycles because of losses in the induction coil and the energy dissipated by the spark at the plug.  This is called a "damped" or "ringing" oscillation.

It's interesting to note that, as the voltage across the points first rises, there is a little "glitch" that is caused because the points haven't opened quite fast enough to keep a small (non-damaging) arc from occuring.  On the above scope picture, the arc occurs at about 150 Volts and only 30 millionths of a sceond after the points started to open.  Since it takes about 20,000 Volts to jump an inch in dry air, I did a little calculation that showed that the points had opened about seven thousandths of an inch when the arc occured.  I suppose the reason another arc didn't occur later was because the voltage across the points was diminished by the arc and, because the points were still opening, the voltage didn't have time to rise to a high enough value before starting the second half of the cycle.  This effect also occurs in the two pictures below and I could see it in every picture I took.

Here are a couple more pictures, this time with the scope only showing the voltage across the condenser.


In the left picture, the usual 0.22 microfarad condenser is across the points.  In the right picture, I've put another condenser across the points.  The period of the wave in the left picture is 130 microseconds giving an oscillation frequency of about 7700 cycles per second.  The right picture shows that adding more capacitance increases the period of the wave to about 180 microseconds giving an oscillation frequency of about 5500 cycles per second.

Let's think a little about resonance.  I suppose the easiest way to picture it is to use a regular bell.  When you give a particular bell a whack, the surface vibrates or moves back and forth at a more or less a fixed rate.  This rate is determined by how fast the surface can spring back after being deformed.  The bigger the bell, the slower its surface moves back and forth and the lower the pitch of its ring.  The smaller the bell, the faster its surface moves and this gives a higher pitched ring.  The thickness and type of metal in the bell also has an effect on the tone.

Electrical resonance works about the same.  As a matter of fact, the sound waves of a perfect bell will look like the above pictures.  What affects electrical resonance is inductance, capacitance and resistance.  Inductance is the property of a coil that makes it resist a change in current.  Capacitance is the property of a condenser that makes it resist a change in voltage.  The circuit will "ring" at a predictable rate depending on the amount of inductance and capacitance in the circuit.  When a circuit is in resonance or "ringing", it's efficiency is maximized.  In other words, a little "tap" will make it ring loudly.

Now, the reason a high-voltage induction coil can produce such a high voltage step-up is due to resonance in both the primary winding AND the resonance in the secondary winding being at roughly the same frequency.  The capacitance across the secondary needs to be only a very small value for it to resonate.  There is ample "parasitic" capacitance in the secondary winding to do the job without having to add any more.   

Now, going to the extreme and leaving out the condenser entirely.....

No condenser!

If there is no condenser in the circuit, the damped oscillation is at around 1 million cycles per second.  Note that the voltage rise across the points is extremely fast and the first couple of cycles have glitches in them.  I think the glitches are caused by multiple arcs across the points.  After about two cycles of the oscillation, the points have had time to open enough (and the voltage has decayed enough) so an arc cannot start.  Now you'd think that, without a condenser, there would just be one pulse and no oscillation until you take into consideration that there is always some "parasitic" capacitance that is caused by the proximity of the turns and layers of the windings.  This parasitic capacitance makes all inductors to "ring" to some extent and, in radio, even this very small capacitance is a detriment and must be accounted for.

With no condenser across the primary of the high-voltage induction coil, the resonant frequency of the primary winding is a LOT higher than that of the secondary winding, so the coil is very inefficient (very little energy transfer between primary and secondary) and produces a very small output.

And, here's a drawing of a buzzcoil.  Note the similarity to the regular Point/Condenser/Coil setup.

All high-tension induction coils are virtually the same.  Buzz coils are essentially all of the above but have what is called a "trembler" that interrupts the current in the primary circuit the same way the points do in the more modern point/condenser/coil systems.  The buzz coil uses the magnetism generated by the primary winding to pull a steel vane toward the core.  Attached to the vane is the movable point that makes contact with the other point that is connected to the battery.

As soon as the magnetism builds up enough to pull the vane, the points open and the field collapses.  With no magnetism to hold the vane, it moves back to it's rest position and the points close, starting the whole operation again.

The only difference between a point/coil system and a buzz coil system is that the buzz coil makes a string of sparks once the battery circuit is completed.  With the point/coil system, you only get one spark every time the points open.

Now, to get right down to it, you actually get several sparks every time either coil "fires".  This is because the oscillation in the secondary makes a spark every time the voltage gets high enough to jump the gap in the plug.

Oh, yes...........Low-tension (ignitor) ignition also uses an induction coil.  This time, though, there is only a primary winding and no condenser is used.  The spark is nice and hot for the same reason the points burn when there's no condenser in a high-tension induction coil setup.

Low-tension Ignitor Setup.

For a hit and miss engine, if you want to put a "spark saver" in this circuit, just break the negative battery line and connect the negative terminal of the battery to an insulated contact that rests against the rocker arm when the exhaust valve is closed.  When the governor latches-up, the battery negative terminal will not connect to the engine frame through the rocker arm and no spark will occur.

In case you're interested, here's what my test setup looked like.

Comments are always welcome.

[email protected]