Conventional Ignition Systems

The historical perspective that led to the T-Type’s ignition system is worth revealing, due to the involvement of some pioneering characters. One such individual was Nikola Tesla (1856 to 1943), born in Serbia during a heavy electrical thunderstorm, in which the midwife was heard to grumble that nothing good would come of any child being born amongst all the lightning flashes. Little did she know that “lightning” was to have such a profound association with Tesla.

Tesla emigrated to the USA aged 28 and worked for Thomas Edison as an electrical engineer, but left when Edison failed to honour his pledge to reward Tesla’s many improvements to DC motors and generators. Tesla joined forces with Westinghouse to develop the AC system of generation and distribution that we all use today.

Renowned for his achievements and showmanship, he became the archetypal “Mad Scientist” with 100s of patents covering such things as wireless communications (and we all thought it was Marconi), fluorescent lighting, poly-phase electric motors and generators, spiral flow turbines and high pressure water cutting. Amongst his many pioneering works, he used the inherent characteristics of inductors to generate very high voltages, enabling him to create artificial “lightning”. The unit of magnetic flux density is named in his honour, as is the Tesla coil, which forms the basis of our ignition coil.

Any coil of insulated wire develops an extraordinary characteristic called inductance, which is enhanced if the coil is wound around a soft iron core. Any current passing through the coil produces an electro-magnetic field, which is a form of energy storage. Changes in current alter the strength of the magnetic field in such a way that the resulting changing magnetic field cutting through the coil’s turns induces a voltage that’s in opposition to the voltage driving the current through the coil. Not easy to find an analogy, but a bit like walking through a non-Newtonian liquid such as custard; the faster you move, the greater the opposition. This opposing voltage is called the back EMF, (Electro-Motive Force), and just as with the custard analogy, the greater the rate of change of the magnetic field, the greater the back EMF.

The overall effect of the back EMF is to oppose any change in current, whether it’s increasing or decreasing, and the energy being stored in the magnetic field is available for this task.

In the case of the TC ignition coil, its inductance has a value of 10 mH (milli Henrys) (where a Henry is the unit of inductance) and a resistance of 4 ohms, so that when 12 volts are applied to the coil, the current takes a finite time to build up to its final value of 12/4 = 3 amps due to the opposing effect of the back EMF as energy is being stored in setting up the magnetic field (Graph 1).

The build-up of current is not linear because the actual voltage applied to the coil’s inductive component is progressively reduced as the voltage developed across the coil’s resistive component increases, (Graph 2).

The characteristic of inductors to oppose change gives rise to a significant effect when the current is turned off. As the switch (contact-breaker points) attempts to disconnect the current flowing through the inductor, the energy stored in the collapsing magnetic field induces a voltage that tries to maintain the same current prior to switching off. This back EMF voltage can reach many thousands of volts and results in serious arcing across the switch contacts as they open and all the energy stored in the magnetic field is dissipated in the arc, (Fig.1).

It was this phenomenon that enabled Tesla to generate arcs many feet long in his laboratory. However, whilst the arcing was impressive, the switch contacts would burn-out unless made very substantial. The back EMF generated was difficult to predict and if too high could cause the insulation of the coil’s wire to break down; so whilst Tesla’s coils were very effective in generating high voltages, their reliability proved challenging.

It’s at this point that Charles F Kettering came up with a simple but ingenious solution in 1908 at the age of 34. Kettering was founder of DELCO and head of research at General Motors and earned himself a reputation for his various aphorisms, one of which shows his determination. He quotes “Failing intelligently is one of the greatest starts as one should keep trying and failing until one learns what will work”.

He is credited with inventing the starter motor after learning of the death from septicaemia of someone injured by the kick-back of a starting handle. The victim had gone to the aid of a lady, who had stalled her car and in the ensuing turmoil had forgotten to retard the ignition timing prior to “swinging the handle”.

This must have been the golden age for inventors, given the many pioneering developments in both electrical and mechanical engineering, together with scientific advancements. Charles Kettering made many contributions, being credited with developing spark plugs, leaded petrol, automatic transmissions, safety glass, cellulose paint, the refrigerant Freon and advanced diesel engines and also held 100s of patents.

Magnetos initially provided the Extra High Tension (EHT) needed by ignition systems based on the magneto effect discovered by Michael Faraday around 1831. Kettering’s attention towards an alternative ignition system to the magneto may have arisen due to its main disadvantage of a low output at slow engine speeds. This was overcome in part by “impulse coupling” which used the release of spring tension to increase the rate of rotation of a magnetic rotor at a crucial moment. The early magnetos used simple cam operated contact points, which suffered from arcing, hence were prone to “burn out”.

Kettering devised an ignition system that used an auxiliary battery to supply current through an inductive coil. The system still needed contact-breaker points to interrupt the current and allow

the generation of a high voltage back EMF. However, he overcame the inherent problems of arcing at the points by the ingeniously simple method of placing a capacitor across the points.

A capacitor is a form of electro-static energy storage consisting of two conductive plates separated by an insulating dielectric. Energy is stored in the dielectric by the reaction of atoms to the stress of an external electric field. (Fig. 2) shows a typical capacitor construction, the two foils forming the conductive plates are separated by a thin dielectric layer and wrapped around to form a “Swiss roll”. Each foil extends beyond the insulator at one end to enable a continuous electrical connection to be made along its edge.

Capacitors work in the opposite way to inductors as they readily absorb current to build up an energy charge between the plates. Once the voltage across the plates reaches the supply voltage, current ceases to flow. However, any change in the supply voltage is opposed by releasing or absorbing a charge of energy. The effectiveness of this energy transfer is limited by any series resistance in the circuit.

Kettering’s original circuit, (Fig.3), possibly drawn on the back of a used envelope was dated July 23 1908 together with his signature and those of two witnesses. It showed an interesting insight as an air gap was included in the soft iron core’s circuit. It’s this air gap that actually stores the energy of the magnetic field.

Whilst the points are closed, the capacitor is shorted out and current is allowed to flow through the coil. As the points open, the dc path for the coil’s current is interrupted but an alternative path for oscillating currents is opened up through the capacitor.

When inductance and capacitance are connected together, they form a resonating tuned circuit, where energy can be transferred back and forth between an electro-magnetic field and an electro-static field. Such an energy transfer produces a sine-wave oscillation resulting from a fundamentally natural phenomenon known as Simple Harmonic Motion, which is also found in other energy oscillating systems such as swinging pendulums and vibrating springs. As a result of this resonance the duration of the spark is extended, giving reliable ignition of the fuel/air mixture in the combustion chamber.

You may be wondering if you haven’t dozed off, why this seems like a physics text book, and you’re not alone – even my wife sympathises with you. However, I’ve come across numerous explanations about ignition systems and found almost as many mis-conceptions, most as a result of an inadequate understanding of basic principles, which, whilst relatively simple, have a profound effect at some quite subtle levels.

As the points open in Fig. 3, the magnetic field collapses, generating a back EMF that charges up the capacitor. When the magnetic field has finally collapsed, the capacitor is fully charged and then starts to discharge back into the coil. This oscillating effect could go on indefinitely were it not for various energy losses. The energy transfer between coil and capacitor offers the chance to control the maximum back EMF generated by the coil as the energy e in Joules within the system can be deduced by two simple formulae. The Joule is a quite small unit of energy equal to 0.24 calories, one calorie being the amount of heat needed to raise the temperature of 1 gram of water by just 1 deg. Centigrade.

For an inductance, the energy e = ½ L x (I x I). where L is in Henries.

For a capacitor, the energy e= ½ C x (V x V). where C is in Farads.

For a typical TC coil of 10mH running at a current of 3 Amps, the stored energy is

e = ½ x 0.01 x (3 x 3), = 0.005 x 9, = 45mJ.

When this energy is transferred into the capacitor across the points, we can work out the maximum voltage appearing across the capacitor, whose value is around 0.22uF.

From e = ½ C x (V x V),

We transpose to find the voltage (V x V) = 0.045/ ½ x 0.22 x 10\(-6) = 409,090.

The square root of 409,090 is 640 volts.

This gives us the theoretical peak voltage across both the capacitor and coil. The value 640 volts is on the high side as the TC ignition coil has a much higher inductance than is usual for modern coils today and therefore stores more energy. A peak output of 300 volts is typical of more recent coils.

Although we refer to the “ignition coil”, it’s really a transformer. The “coil” we’ve been mentioning forms the primary winding of a transformer, whose secondary winding contains many times more turns, so the magnetic flux generated by the oscillating primary current also cuts through the numerous secondary turns and consequently induces a much higher voltage of about 25,000 volts, (Fig. 4 ).

So if we require 25,000 volts at the secondary, we need a turns ratio of 25,000/640 = approx. 40 to 1. Newer coils of an approx. 5mH inductance giving 300 volts need a higher turns ratio of 80 to 1.

As we’ve seen, the capacitor is able to control the peak voltage across the primary of the ignition coil, but it has one more important function. Without the capacitor, the back EMF generated across the opening points would instantly reach a sufficiently high voltage to force the current to jump across the

point by arcing. This would not only erode the points but would dissipate much of the coil’s energy in the arc. The capacitor slows down the rate at which the back EMF can rise by absorbing some of the energy of the collapsing magnetic field in such a way as to produce a sinusoidal waveform thanks to the resonant tuned circuit formed by the coil and capacitor.

The rate at which the oscillating back EMF rises is never so great as to initiate arcing across the opening points, (see Fig. 5a), and consequently preserves the life of the points. However, the capacitor has a stressful life coping with oscillating high currents at elevated temperatures, which can lead to a partial failure that may still produce a spark at a spark-plug in the “open”. The spark will be weak and put under pressure can fail. Such a condition occurs when the internal series resistance of the capacitor increases to a point where it absorbs too much energy and/or causes arcing at the points (see Fig 5b) and results in either erratic running of the engine or a failure to run at all. Another possible failure arises if the insulation of the capacitor’s dielectric becomes “leaky” or even “shorts out” (see Fig. 6) this can be measured with an ohmmeter, unlike the series resistance, which needs a special bridge.

New “old stock” capacitors can show alarmingly high series resistances. Out of 10 tested, Rs varied between 9 and 140 ohms, the average being 44 ohms. Modern replacement capacitors had an average Rs of 4 ohms, which is a harbinger of problems to come, as such high resistances indicate poor quality in both design and manufacture.

Finding an alternative capacitor that could be substituted for the soldered in capacitor used on the dizzy plate seems a worthwhile endeavour.

The Quest for the wholesome capacitor.

Lucas quote the capacitor value to be between 0.18uF and 0.24uf, so the preferred value of 0.22uF fits in well. The choice of a low loss dielectric capable of handling peak voltages of over 600 volts, high currents and high temperatures was narrowed down to polypropylene, a dielectric that’s been reliably used in high-current pulse-discharge circuits due to its low Rs and robust dielectric.. Two versions of the capacitor are suitable; the first can be housed in a thin copper tube soldered to the dizzy plate and retained in the distributor, the second is housed in some 22mm copper plumber’s fittings and fixed next to the coil with a flying lead going to the CB terminal. This is an improved position as the coil and capacitor form a tuned circuit, (see Fig.7).

Version 1 uses a Vishay MKP 1845, 0.22uF rated at 1,000 volts, +/- 5%, and up to 100 deg.C operation. 44mm long by 18mm dia. and is suitable for the “soldered-in” type of capacitor, (see photo 1).

Photo 1

Version 2 uses a LCR PC/HV/S/ WF, same characteristics but +/- 20%, 33mm long by 20mm dia. This is more suitable for attachment to the side of the coil in those cases where the dizzy capacitor was secured by a screw to the dizzy plate. (see photo 2).

Photo 2

Both need to be potted in araldite with a short flying lead made from multi-stranded silicon sleeved cable rated at 1.5Kv, -60deg.C to +180deg.C.

Photo 3

So why are both original and modern capacitors unreliable? The first clue came from being able to vary the series resistance Rs from 300 to 6 ohms simply by pressing the rubber end seal on a faulty modern capacitor. Dismantling the capacitor revealed the rather crude construction shown in photo 3, where electrical continuity relies on pressure exerted by a rubber washer. This presses a contact-plate against the end of the capacitor element, which in turn is pressed against the end of the housing. This form of construction is suspect due to:-

1, The elasticity of rubber is reduced by exposure to heat, oil and time.

2, Electrical contact is made only at the “high spots”.

3, Any residual solder flux will act as an insulator, unless cleaned off.

4, Contact materials are of dissimilar metals.

5, Access for contamination.

These points indicate sub-standard design and construction compared with an industrial quality capacitor, where the leads are electrically u-welded by a plasma-arc spray or a vacuum deposited aluminium film on the conductive foils protruding at the ends of the capacitor element, photo 4. The whole assembly would then be potted with resin inside a case to provide a permanent seal.

Photo 4

Present replacement dizzy capacitors cost about £7 for a “screw-on” type and some £40 for a replacement “soldered-on” type, as compared with the cost of polypropylene capacitors at about £2.50 from a trade supplier or £4 on e-Bay. However, the time and effort needed to install polypropylene capacitors in a metal sleeve, attach leads and pot in araldite would seem to make them commercially uneconomical unless constructed on an industrial scale.

It’s probably only when the second replacement dizzy capacitor causes erratic running or brings the engine to a halt that many will take an interest in this issue. Capacitors in a partial failure mode may well affect the life of contact points, so problems due to premature erosion of the points should trigger suspicions about the capacitor.

If there’s sufficient interest, a follow up article on how to fabricate replacement capacitor modules may be considered, but in the meantime any feedback is welcome.

Eric Worpe

10 thoughts on “Conventional Ignition Systems

  1. Joe Policastro says:

    Thank you for this great article. Please do not leave us hanging. We still need to fix the problem. Now we know why the new ones fail out of the box. Please tell this owner how to fix it.

  2. Neville Oates says:

    Dear John, I enjoy your TTT-2 publication a great deal and as an MGTC “freak” for some time and currently the editor of a local car club newsletter not MG related but nevertheless with quite a number of MG’s in the club I often hear the comment, Oh! not another MG in the club. As it happens I am always looking for interesting and technical articles to put in the newsletter, and I would ask your permission to reproduce the article on coils and electrics in general. Apart from anything there is a wealth of history in the article and I know many of us older members like to learn a little more. Hoping that this is OK, regards Neville…keep up the good work

    • JOHN JAMES says:

      Dear Neville, Thanks for your comments. I have checked with Eric and he is content for you to reproduce his article as long as appropriate accreditation is given.

  3. Neville Oates says:

    Great article Eric, a problem all of us encounter at one time or other, capacitors can be the bain of our tinkering with old cars. Yes I would be interested in a follow up article

    Cheers Neville Oates

  4. Sam Stephens says:

    Dear Eric,

    Thank you for your brilliant article on condenser history. I think I need a little lie down now after reading it.

    I have just experienced a “new, out of the box” failure of a condenser on my TC (and what a game it has been to realise what the problem was). Anyhow, I am about to install yet another new condenser in the hope of some better luck. Do you have an opinion about the reliability and effectiveness of these new “electronic, 2-way” ignition systems that are supposed to be OK for T type period engines?

    • Eric Worpe says:

      Dear Sam,
      Thanks for your kind words, I also has several lie downs whilst writing it.
      Electronic ignition systems usually perform very well. However, the TC appeals to me because of its basic simplicity and to make it more complicated by installing electronic ignition might detract from being able to fix the ignition system should a problem occur. I spend my working life fixing electronic problems so that simple systems have great appeal.

  5. Richard Howell says:

    Eric I have installed a Boyer Bransden inductive discharge unit. This means disconnecting the capacitor and changing the route of the wires coming from the coil. The points still work but do not burn so much. Is this stepping away too much from originallty?

  6. John Penny says:

    Eric, Many thanks for the article. Having experienced problems with modern replacement condensers and resorted to replacing the points with “Hall effect” electronic triggers such as the Powerspark units, I would be interested in a follow up article on how to make reliable replacements. I too think that keeping it simple is part of the attraction and the ideal solution for our old cars.

  7. Kelso Riddell says:

    An excellent paper. I have a Polski Fiat 126P out here in Cuba, and just replaced two condensers, both new old stock original Magneti Marelli. I was astounded today to have to replace one after a week. Now I fully understand and will try to obtain parts to build a replacement. There will be some delay as there are few routes for getting stuff into Cuba at present. I just have to hope the one I have fitted will last three months or so. Otherwise I will strip down the old ones and see what the resourceful engineers here can do.

Comments are closed.