We don't believe our potential customers are totally ignorant on the subject of their car's functioning. Nevertheless we realize that better informed customers can tell us clearly what they want and what they should expect but that it often difficult for them to get objective information. By making this article freely available to you we hope to give you some extra knowledge on the functioning and importance of ignition systems of gasoline powered engines so that you know which questions to ask and what to do when you want to improve your car's performance and economy by upgrading the ignition system. Where necessary we have changed some words to make the article more understandable for Europeans with regards to expressions for fuel consumption and engineering units.

Article copied with permission of Jacobs Electronics@ from C.Jacobs recent book, "The Doctor's Step-By-Step Guide to Optimizing your ignition" available at CDC.

The insiders theory on ignition

If you want to appear really knowledgeable about the effects of ignition on fuel economy and performance, all you have to do is took real pensive, stare off into the distance and say, "Well, it really isn't quite that simple." Say this no matter what ignition fact the other person states and you'll always be right because ignition isn't that simple.

Here are some examples:

Other person:"More spark energy gives more power.
You:"Well, it really isn't quite that simple."
Other person:"Higher octane fuel lets you advance the timing for better performance "
You'll be right again if you say, "Well, it really isn't quite that simple." See what I mean?

In this section, I'll discuss what is known about the fundamental theory of spark generation, spark properties, and timing requirements, and show how significantly ignition requirements change when going from a stock to a modified, off-road application, or just re-jetting your carburetor. I will show how ignition affects mileage and performance and provide some solutions to technical problems. More extensive technical solutions are discussed in later chapters.
As a rule, people give more time and attention to things they can see than to invisible things. Since we can't see electricity, ignition is probably the least studied and one of the most controversial of all automotive technologies.
So visually oriented are people in general, and Americans in particular, that
I'm repeatedly asked, "How important can ignition be? It's just a "simple coil and battery and things like that." My standard comeback is to jokingly suggest they perform this "scientific" test: I tell them, "Next time you're on the highway and it's safe to do so, reach over, turn off the key, and see if you notice any difference at all in the way the engine runs." (I don't really recommend you do this because of locking steering wheels, and so on, but it sure gets the point across.)
What little ignition research is done is primarily in the area of spark timing which can be "seen" with a timing light Only very slight research has been under taken in the area of spark quality (which) deals with spark duration, intensity, and so forth), and yet all the research done Jacobs' Electronics laboratories proves that (not "quite this simple") poor spark quality is robbing more fuel economy and performance from people than is poor timing.

HOW DOES AN IGNITION SYSTEM GENERATE A PROPERLY TIMED SPARK?

Except for jet turbine engines, which burn continuously, most engines in general, and reciprocating, as well as rotary engines in particular, require that fuel be continuously reignited at precisely timed intervals. In properly operating production automotive engines, fuel that is burning at any instant in time will deliver its mechanical energy via some sort of crankshaft, then pass out the exhaust and not be used to ignite the new incoming fuel. Engineers must design and specify some means to assure the new incoming fuel will be repeatedly ignited (about 150 times each second) in such a way and at such time as to convert the largest percentage of the fuel's chemical energy into mechanical energy. In a modern automotive engine, typically 15% of the chemical energy stored in the fuel is so transformed. This percentage is referred to as the fuel efficiency.
This fuel efficiency is so low that a loss or gain of just three percentage points could make a 20% difference in your fuel consumption. A properly working ignition could certainly make a three-point shift in fuel efficiency when compared to one that is either not igniting properly or not igniting at the proper time.
To convert the fuel's chemical energy into mechanical energy, an internal combustion engine needs six things to happen:

1. Air and fuel must be brought together in the right ratio.
2. Fuel and air must be steered into and out of the right location.
   This location is typically called the combustion chamber. Steering this air and fuel is usually provided by valves and cams.
3. Fuel and air must be prepared to make them burn most efficiently.
   An example of this preparation would be engine compression.
4. Air and fuel must be ignited at the right time so that as they burn and thereby expand, chemical energy is converted.
5. Means must be provided to convert the burning-induced increase in gaseous pressure into mechanical energy. This is typically done by pistons, connecting rods and crankshafts.
6. Burned air and fuel must be eliminated so the new air and fuel can enter -- exhaust.

This section, and in fact this entire book, is primarily concerned with item 4, ignition, and with only one means of igniting fuel, electric power. Therefore, whenever I use the word "ignition," it will mean ignition caused by electric power. However, there are other means. For example, a diesel ignites fuel by compressing air until it gets so hot, it's above the fuel's ignition temperature. Then it simply brings the fuel in contact with this heated air and the fuel ignites due to the air's high temperature.
Modern production ignition systems are usually comprised of eight components or functional units, which must work together. Each one of these functional units has its own section in this book, which explains not only its design parameters, but also gives criteria to help you select the best type for your application and budget. This section discusses how they work together to form the functional unit we call the ignition system.
The most well-known of all ignition functional units is probably the sparkplug. Engineers refer to this as a transducer, which is any device that converts one form of energy into another. A car speaker is a transducer in that it converts electric energy into sound energy.
The most commonly constructed sparkplugs have a center electrode, usually made of metal, which is electronically isolated and separate from a side electrode. This separation is commonly called the sparkplug's gap, which can range from .018 in. to .125 in. (0.46 to 3.17 mm) in a modern production engine.

It is an established law of physics that if sufficient voltage, in the neighborhood of 32,000 volts per inch in free air, is impressed across a gap, a spark will arc or jump the gap. A spark is no more than current flowing through air, or in the case of an engine, through a mixture of air and fuel. Generally, sparks can be made to jump at a lower voltage when arcing from sharp objects. This is why new sparkplugs, with electrodes terminated cleanly, will require less voltage to arc than ones with electrodes worn round. Also, less voltage is required when negative voltage is on the sharp tip. Contrary to what appears on engine diagnostic oscilloscopes, in virtually all production engines it is the negative voltage which is applied to the center electrode. Engine scopes automatically reverse the polarity on their screens. That is, the negative is shown as up; positive, down.
If fuel is present in the gap, it is not fully understood why this current will cause it to start burning. The possible reasons are discussed elsewhere in this book, but for now, it is sufficient to say that under the proper conditions a spark will ignite the fuel most of the time (remember, it's "not that simple"). Most sparkplugs use a sparkgap of some sort to ignite fuel, but there are sparkplugs available which use solid state ceramics to do the same job. While they work well, they are not that popular because they tend to be fragile.
The sparkplug itself has little to do with timing, which is controlled by other means. The sparkplug immediately transduces the incoming high voltage into a spark.
When a sparkplug is properly installed in an engine, its combustion chamber gap location is very important. The gap should be in such a position that when the spark jumps the gap, and if a flame is ignited, this flame will start in the most efficient location. Efficient means that as it propagates, it will convert the fuel present into the maximum mechanical energy. To a certain extent the size of the gap is a simple extension of this search for the optimum gap position because burning begins over virtually the entire length of the spark gap.
In theory, it would be better to have the high voltage generator and sparkplug center electrode as one unit, but for purely practical reasons of heat, vibration, size and cost, the generator of high voltage is almost always located at some distance from the sparkplug or plugs on multicylindered engines.
The most common conduit of this high voltage is sparkplug wires which engineers usually refer to as high tension leads. A perfect high tension lead would conduct all this energy from the high voltage generator to the center electrode. Unfortunately nothing made by man is perfect, and there are many types of sparkplug wires which in some way lose a part of this spark energy. However, regardless of type and other forced restraints, such as reduction of radio noise, all sparkplug wires are simply trying to pick up high voltage energy from one location, usually but not always a distributor cap, and deliver it to another location, the sparkplug center electrode. The distributor cap and rotor, like the ignition wires, would be eliminated if you had the theoretical optimum of the high voltage generator and sparkplug center electrode as one unit. A distributor rotor and cap allow for the economic advantage of having one high voltage generator provide spark energy for all the cylinders.

In a four-cycle engine, which is by far the most common (intake, compression, power, exhaust), a spark is needed on only every other rotation of the crankshaft. For this reason, the distributor shaft, to which the rotor is attached, is geared to only rotate at half crankshaft rpm.
In the traditional and still most common rotor and cap assembly an electronically-conductive rotor is secured to the end of this one-half engine rpm distributor shaft. One end of the rotor's conductor is directly over the center of the distributor shaft and the other end extends outward to almost the radius of the distributor cap.
The rotor and shaft have the same spatial relationship as does a foot (rotor) to a leg (distributor shaft). Rotation of the shaft would be equivalent to spinning around on your heel, which is planted in one location on the floor, and your toes would then be executing a circle of 360 degrees.
Continuing with this analogy, if you drove nails into the floor, one for each cylinder, just far enough outside the circle described by your rotating toes such that the toes would just brush but not hit them, then the floor would in effect become your distributor cap and the nails the distributor cap terminals.
The high voltage generator is connected to the rotor (heel of the foot) via a contact rotation button and spark energy is conducted along the rotor (foot from heel to toe) to the distributor terminals (nails in the floor). From there energy is conducted via sparkplug wires to each sparkplug center electrode.
While this is generally the way it works, there are exceptions. Ford, in the early 80's, with their EEC-IV, used a two-level rotor-the equivalent of having one foot on top of another while spinning in the same direction. In the mid-80's, Buick introduced some systems (called DIS, Distributorless Ignition Systems), that
incorporated multiple high-voltage generators and no distributor cap and rotor. In virtually all modern engines high voltage generation is accomplished using some form of electromagnetic waves much like radio waves. That's one reason why ignition can interfere with and cause noise on your radio. Two laws of magnetics are employed to generate high voltage. One says that if you take a wire and bend it into loops forming coils, and then take a common magnet, like the type that holds papers to a refrigerator, and move this magnet near the coils, you'll get the same voltage generated in each and every coil. Since the voltages are equal and in the same direction, voltage goes up in direct proportion to the number of coils.
Also, the faster you move the magnet the proportionately higher the voltage. Double the speed, you double the voltage. The direction of travel affects voltage polarity. That is, if moving the magnet toward the center of the coils causes a positive voltage, then pulling it away would cause a negative voltage.
Based on this law, you could generate high voltage by wrapping a wire into multiple concentric coils, embedding a magnet into the tip of a high-velocity bullet, and then firing the bullet near the center of these coils. As the bullet approaches, the voltage would be in one direction; as it passes and exits, the voltage polarity would reverse.

Every time you wanted a spark you'd have to fire another bullet. In a V-8 wound up tight, say 7,500 rpm, you'd be firing 30,000 bullets per minute! Well, unfortunately for the sales of Winchester and Remington, but fortunately for us, there is another law of nature that can trick these coils into thinking you're moving a magnet near them, even though you are not.

If another set of coils is taken and a steady current is forced into them, they will generate a facsimile of the permanent magnet. If this first set of currentcarrying coils is nearby, the second set, the ones generating the high voltage, won't know the difference between the action of the first set of coils and a permanent magnet. That is, if you increase and decrease the current in the first set of coils, the second set will act as if you are moving the magnet around. As you increase the current, it will cause the same reaction as if you moved the magnet closer. The faster you increase the current, the faster it acts as if the magnet is being moved in closer. Conversely, as you decrease the current, it feels as if you're moving the magnet away.
If you wrap the first and second set of windings around a piece of magnetically influenced metal, such as laminated steel, this magnetic interaction or coupling can be enhanced or made to appear as if the windings are even closer together.
That is basically how a modern highvoltage generator works. The metal enhancer is called the core. How much it enhances is called its permeability. The first set of wires, the ones in which the current is actively increased and decreased, is called the primary. The second set of windings, which usually have about one hundred times the number of turns of the primary, and actually generates the high voltage, is called the secondary. This entire functional unit is called the coil.
In practical common induction-type ignitions, the current is rapidly changed in the coil primary, simulating rapid magnet movement. This happens every time a spark is required. The output of the high voltage secondary is connected to the shaft end of the rotor, equivalent to the heel of the spinning foot. From there the high voltage is distributed to each cylinder.
The rest of the ignition system is concerned with getting this primary current to rapidly change, what engineers call rapid magnetic flux change, just at the time a spark is needed. By controlling this time (timing), engineers are trying to achieve the same objective as with spark gap location, that is, start the fuel burning at that piston position that will achieve the highest percentage of conversion of the fuel's chemical energy into mechanical energy. To do this, the primary current is rapidly changed, i.e., the spark is optimally timed by knowing the position of the piston.
When the piston is in the optimum position, the primary current is interrupted.
There are two common methods of rapidly changing primary current: the deceleration method, which is the most popular, and the acceleration method. Remember, except for polarity, the secondary doesn't care which is used and the polarity is handled by simply reversing which end of the secondary goes to the sparkplug center electrode.
To understand the deceleration method, picture that the primary coil windings are connected between the plus and minus terminals of the vehicle's battery. Due to the physical properties of winding primary wire around a piece of metal, the current inherently increases so slowly that no spark can be generated. Eventually the current will level off at a value set by the voltage of the battery balanced out against the accumulated electrical resistances of all the wires and any resistances placed there on purpose. A resistor placed there intentionally is usually called a ballast resistor. Its job is to reduce the value to which the coil primary current finally comes to rest.
Once this equilibrium is reached, if you then cut one of the coil's connection wires, typically the one to the minus battery terminal, the current in the primary would obviously decelerate very rapidly, generating the high voltage required to make a spark. The objective, of course, would be to time the cut precisely to when the piston is in the optimum position.
A modern inductive ignition does all of this and more using the deceleration method of rapidly changing primary current. The coil primary is connected to the positive battery terminal via a switch activated by the ignition key. This end of the coil primary is usually called BAT or (+). The other end, usually labeled (-), goes to a switch which repeatedly takes the place of cutting the wire. This switch is usually an NPN transistor, where the Emitter is connected to ground, and the Collector is connected to the (-) coil terminal.
However, for purposes of illustration, we will use a more visually easy-to-understand set of physical metal contact points. One contact is connected back to the negative battery terminal. The other contact is the one that goes to the (-) side of the coil primary. When the key is turned on and if this switch is closed, the current path is from the (+) battery terminal, through the key switch, to the (+) side of the coil primary. It then goes in through the coil primary out the (-) side to one contact of the points switch. Call this the (+) contact (Collector). It then proceeds through the (+) switch contact to the (-) switch contact (Emitter), and finally back to the negative battery terminal. When the switch contacts open, coil primary current rapidly decelerates and a spark is generated.
Until 1972, the switch was virtually always a set of contact points, commonly called points. The negative contact, connected to the battery terminal, was called the fixed point because it was secured to a good ground and never moved in normal operation.
The other (+) contact, connected to the (-) coil primary, was called the movable point. To interrupt the primary, it was physically pushed away from the fixed point by a part called the distributor cam. It is common to put a capacitor, commonly called a condenser, across the points to allow the contacts to break clean. It was also installed because coil primary current, when rapidly interrupted, and without a condenser for protection, would electrolize and corrode the metal of the contact points. Even with the condenser, points erode, but at a slower rate. Fortunately, NPN transistors don't need this big condenser.

In 1972, Chrysler Corporation started selling vehicles with a concept in induction ignitions that had been around for years, which subsequently came to be known as electronic ignition. Today virtually all OEMs, including General Motors HEI, use this same concept. While it utilized the same principle and connections as the contact point inductive system, the actual points were replaced with an NPN type transistor, described in part above, which has three terminals: emitter, base, and collector. The transistor used in most modern electronic ignitions is an NPN.
The way a transistor works is that when a small amount of current is supplied into the third transistor terminal, the base, its collector and emitter appear as if they were closed contact points, and current builds up in the coil primary. When this small current is not provided into the base, the collector and emitter quickly appear as open contacts, interrupting coil primary current just as contact points used to.

Capacitive discharge, CD, and certain types of magnetos work in the opposite way. They start with zero primary coil current, then rapidly accelerate it when a spark is desired. This sudden current acceleration generates high voltage in the secondary. That is why these functional systems are known as the acceleration type. Either with CD or magneto, one side of the coil primary is generally grounded as opposed to connected to the (+) battery. When a spark is required, a mid-range voltage of 250 to 600 volts is quickly connected to the non-grounded end of the coil primary. This mid-range voltage is so high, and the coil primary current acceleration so rapid, that high voltage is generated in the coil secondary. In the case of the CD, a capacitor is slowly charged to, say, 480 volts. When the points, either mechanical or transistor, open, this opening is sensed, and switching means inside the CD box cause the charged capacitor to connect quickly across the coil primary.
In common inductive type ignitions only 12 volts are impressed across the coil primary and current builds up slowly, so slowly in fact that it cannot generate a spark. On the other hand, when 480 volts is impressed across the coil primary, 40 times as much voltage, which is 40² = 1600 times as much power, it is quite sufficient to accelerate primary coil current to the point well in excess of that needed to generate high spark voltage in the secondary. That is essentially how a CD system generates a spark.

Magnetos can be one of two types: either the acceleration type, which usually has an external coil and two primaries (Mallory is a manufacturer who primarily makes an acceleration -type magneto); or deceleration type, which usually has the entire mechanism in one housing around the shaft and only one coil primary (Vertex is one manufacturer of deceleration type magnetos). While manufacturers may have different design criteria for their magnetos, there is no inherent spark advantage between the acceleration and deceleration types.
The deceleration type basically works the same as an inductive type ignition with the exception that it has its own little alternator to get current moving in the coil primary. The alternator is comprised of a magnet mounted on what is called the distributor shaft. As the shaft rotates, the tip of the magnet swings around closer, passing by, and then farther away from the windings on the coil primary. As it gets closer, voltage and hence current are gradually building up in the forward direction in the primary. When the magnet has induced the largest possible current in the primary, and the tip of the magnet is starting to move past this peak current point, that very instant is when the points open. In the same way as with induction systems, the points are connected in series with the primary. Therefore, when they open, the primary current rapidly decelerates and high voltage is generated in the secondary.

An acceleration-type magneto has the same magnet and series point arrangement as does the deceleration type. That is, a magnet builds up current in wires which are wrapped around a core and at the peak of current, which is when the spark is desired, the points open. However, in this case there is a second coil primary in parallel with the first. All the current, which had built up in the first primary, is quickly channeled into a second parallel coil. As with an induction ignition coil, this second coil primary has a secondary wrapped around its core. When this current in the second primary rapidly accelerates, it produces high spark voltage.

Not yet available OEM is a superior type called a computer ignition. A computer ignition draws in 12 volts DC and an energy processor converts it into a form of AC which more closely matches the impedance of the coil primary. Impedance is just a complex form of resistance. When any two functional units have matched impedance, electrical information can be transferred from one unit to the other more efficiently and at a higher speed. All ignitions try for matched impedance. The only difference is that computer ignitions have active components to maintain the match in the face of changing engine environment and running conditions.
The output of the energy processor is stored in a resonant buffer, whose purpose it is to take in something fast, usually energy or information, and deliver it slowly; or conversely, take it in slowly and deliver it quickly. This buffer is doing the latter.
A trigger is connected between an electronic switch inside the computer and the vehicle's original ignition, contact point or electronic. When the points or transistor open, the electronic switch closes and the buffer's energy is connected to the spark processor regulator.
The buffer has too much stored energy for normal ignition. If it were all delivered to the coil on each spark cycle it would cause detrimental results, such as excessive plug erosion and spark energy.

The regulator uses a feedback control information technique, and by measuring the changes in spark gap impedance, the coil acts as a transformer, and the sparkplug an "in-cylinder sensor." A computer ignition actively restricts and regulates the amount of spark power and energy, which can pass from the buffer to the coil to assure impedance matching and reliable ignition, but not so much energy as to over-spark.
A day-to-day analogy would be you sitting in control (bass, treble, volume) of a 1,000 watt amplifier and six switches (rotor in a 6-cylinder engine), which connects this amplifier to six different speakers (each cylinder's combustion chainher). In addition, these six speakers are to be moved to many very different size and sound-deadened show halls (changing engine environmental and running conditions).

On the average, you will only need 75 watts with the bass and treble in the middle to fill the hall with the optimum sound. You have a technician in the room calling to tell you how to adjust the amplifier to optimize that speaker in that hall (the computer ignition's ability to use the spark plug as an in-cylinder sensor). With modern aftermarket computer ignitions, not yet available OEM, this optimization process takes less than 1.5 degrees of engine rotation.
The advantage over, say, a fixed 75-watt amplifier, with bass and treble set in the middle, would be that just as each cylinder's spark needs change (due to fuel, temperature, compression, carbon buildup, driver, etc.), so each speaker's sound and efficiency changes, leading to a much sweeter sound (performance, mpg, engine life, time between tune-ups)!
Until this point it was assumed the ignition system knows the piston position. Timing, which engineers call "engine position-controlled switching means" (EPCSM), requires generating a spark in relation to piston location. It is essential that piston position be known if the points or transistor is to generate a spark at the optimum piston position. In production engines the points or transistor know where the piston is because they are mechanically or electronically coupled to the actual crankshaft, either directly via a crank trigger arrangement or coupled through the distributor shaft which, in turn, is geared to the crankshaft.
Going back to the example of a spinning leg and foot where the leg was the distributor shaft, the most common way of telling the EPCSM what position the pistons are in is to simply connect the distributor shaft to the crankshaft via 2 to 1 step-down gears. In that way, by opening or closing the EPCSM in relation to this distributor shaft, you are automatically timing and distributing the spark in direct relation to the piston position.
In a contact point system, a part, which is shaped like the head of a bolt, is concentrically pressed onto the distributor shaft. It is called a distributor cam because the protruding bolt lobes, one for each cylinder, literally get in the way of the movable contact, thereby converting the rotational shaft motion into straight line point motion. Converting rotary motion to straight line motion is called "camming" The distributor cam pushes the movable contact in a straight line away from the fixed contact, opening the circuit and interrupting coil primary current. The net effect is like a chain reaction. That is, the piston position is in fixed relation to the crankshaft position. The crankshaft completely controls the distributor shaft position, which, in turn, rotates the distributor cam.
Therefore, the cam's lobes will interrupt primary coil current by forcing the movable point to open in a fixed spatial relationship to piston position. It is like a chain with multiple links. When you pull on one end, the other end moves the same amount. This generates the spark in fixed relation to piston position.
An electronic ignition uses virtually the same concepts except that a reluctor instead of a cam is keyed around the distributor shaft, and a pickup coil, mounted close to the spinning reluctor, is used instead of the contacts. Shaped like a gear, the reluctor generates voltage as if it were a moving magnet. When the gear tooth rotates first closer, then farther away from the coils of wire on the pickup coil, the coil reacts as if the magnet was moved closer, stopped, then moved farther away. As a reluctor tooth, one for each cylinder, is moving closer to the pickup coil, a positive voltage is induced in the coil. This voltage, via electronics inside the ignition amplifier or module, generates current into the base of the transistor. As the tooth passes by the pickup coil and starts to move away, a negative pickup coil voltage is induced. It is this negative voltage which in turn stops the base current, opens the contact between collector and emitter, and interrupts primary coil current. When this current is so interrupted, it generates a high voltage spark in correctly timed relation to piston position.
A crank trigger mechanism uses this same reluctor/pickup coil principle except it mounts the reluctor directly onto the crankshaft, eliminating the tolerances of distributor gears, shafts and bearings. The reluctors on crank triggers have only one tooth for each two cylinders because, as discussed above for a four-cycle engine, you only need a spark every other time the piston comes up. The pickup coil, mounted in proximity to the rotating crankshaft reluctor teeth, serves the same function of generating plus voltage as the teeth approach, and negative voltage as they leave. With distributor electronic ignitions, the instant voltage goes from positive to negative, the spark is generated. If the pickup coil is properly mounted, it will capture the optimum piston position.
These are the fundamentals of how a spark is generated and timed to the optimum piston position for maximum fuel efficiency. The means by which these functional units work is neither controversial nor simple. What is controversial anxd definitely not simple is determining this optimum piston position. In other words, with an ability to generate a spark at any desired piston position, it becomes necessary to determine what this optimum is.
To make the matters even "less simple" good spark quality requirements are also controversial and continually changing with alterations in engine environmental and running conditions. I don't want to give the impression nothing is known about ignition because you can be quite certain of eight facts, and using these facts can get you better fuel efficiency that is, mileage or power, depending on your objectives.

1. Ignition -- one of the surest improve mileage and power. Ignition is one of the very few engine improvements you can make that will improve mileage and performance. When you change carburetion, cams, piston/crankshaft or even exhaust systems, you may well have to give up performance to get good mileage or vice versa. This is not the case with ignition. For this reason, ignition should be your starting point (no pun intended) when trying to achieve better mileage and performance simultaneously.
   NOTE: I WILL BE USING "MILEAGE", "PERFORMANCE", AND "POWER" INTERCHANGEABLY, BECAUSE WHEN IT COMES TO IGNITION, THESE WORDS ARE SIMPLY LABELS FOR FUEL EFFICIENCY. IT IS COMPLETELY YOUR DECISION WHETHER YOU WANT TO USE THIS IMPROVED FUEL EFFICIENCY FOR BETTER MILEAGE, MORE FULL RACE POWER, OR TO BLEND THEM INTO WHAT HAS COME TO BE CALLED "IMPROVED PERFORMANCE."
   
   
2. Playing the odds in ignition. You can never completely eliminate misfires. The original equipment ignition does it best and has the lowest percentage of misfires: (a) at speeds of 25 to 40 mph (b) using light throttle, and (c) using gasoline as opposed to propane or compressed natural gas (CNG). Because you get the least misfires under those conditions, you also get the best mileage. Likewise, it's at its worst when the engine is cold, or when using fuels that require different spark profiles. From the time you start your cold engine, including time the choke is operating, a well-tuned system gets good, reliable ignition on the average of 88 to 94% of the time on gasoline (6 to 12% misfires with corresponding wasted gasoline). By an average I mean, once the engine is warmed up and driven in a cruise mode, this instantaneous percentage can drop close to 3%, but exceed 12% on cold start and run. These percentages are derived using Champion's new style thermocouple sparkplugs, with ultra-fast-acting heat sensors capable of determining if combustion occurred on each cycle. When misfires are computed from hydrocarbons (HC) in the exhaust gas, the percentage misfires typically appear lower partly because of instrument differences and partly because some of the unburned gasoline ignites in the exhaust manifold.
   
   
3. Lean mixtures-good fuel economy or disaster? Unfortunately, in many cases, the maximum fuel economy that can be achieved by carburetor (fuel injection) or mixer (propane natural gas) improvements is limited by the vehicle's ignition system, not by fuel-burning considerations.
   An air/fuel ratio of 11 parts of air (by weight) to 1 part of most fuels is most conducive to spark ignition. The leaner the mixture the harder it is for an electrical spark to ignite it.
   Even with a choke, the mixture is liable to be inappropriate at cold start. Without some source of heat the gasoline cannot evaporate, and this is exactly when the ignition system is at its worst because a cold battery is struggling to crank an engine full of cold oil. Consequently, a choke mechanism is used to provide this richer, easy-to-ignite mixture. Once the engine is warm, the choke backs off and a somewhat leaner mixture can be provided. Here again, leanness is indirectly set by the limits of what the ignition system can ignite, not by what the engine could most efficiently burn. It is further complicated by the fact that a spark ideal to start a cold engine would lead to bad fuel economy, cavitating, and so forth, once the engine warmed up. When using alternate fuels such as propane and CNG, the radical change in spark requirements becomes even more severe.
   While my testing has indicated you can almost always hurt mileage and economy by using more spark energy than the optimum, there is data to indicate that more ignition energy correctly timed is always better; however, the benefits may not be worth the expense and additional sparkplug erosion. Also, more ignition energy may require less timing advance for best performance due to less delay in establishing a flamefront.
   
   When trying for maximum fuel economy, the operator sets air/fuel ratios by continuing to lean the mixture. In general, the leaner it is, the better the fuel economy up to the point where, even in a hot engine, the mixture is so hard for an electric spark to ignite that the loss due to an increased percentage of misfires outweighs the gain from more efficient burning. To make matters worse, the government insists on butting in with its emission standards. As it turns out, leaner engines reduce HC and CO emissions but normally increase NOx emissions well beyond current regulations except in the case of ultra-lean special engines, such as the Honda stratified charge design. In view of this, almost all current U.S. engines are run richer than would be best for burning efficiency- very near the chemically correct 14.7:1 by means of an air/fuel ratio feedback control or even richer-to reduce NOx emissions. Then they often use an air pump to add additional oxygen to the exhaust system to burn up HC and CO.  Some surging in these vehicles is caused by leaner mixtures and large amounts of exhaust gas recirculation also done to reduce NOx emissions. Improved ignition can, in most cases, reduce this surging by stabilizing the ignition delays and total combustion process. To meet emission standards, many OEM manufacturers are running engines so lean that fuel mileage and vehicle performance are hurt substantially (that is, their percentage of misfires is quite high).
   It used to be (circa 1974-77) that you could achieve better fuel economy in many cases by richening the mixture, thereby lowering the percentage of misfires. Remember, it's easier to ignite a richer mixture. Enter the government again; the newer fuel systems are designed to prevent you from changing air/fuel ratios.
   On the other hand, if you could guarantee ignition almost 100% of the time, it probably would not improve economy because at present the most commonly used emission control strategy is not lean calibration but either stociometric 14.7:1 calibration with three-way catalytic converter and oxygen sensor feedback control, or slightly rich calibration with catalytic converter and air pump to burn up excess HC and CO in the exhaust system. In either case, a richer mixture would not be likely to improve economy once you're guaranteed near 100% ignition. Where present strategy fails is that while they do lower emissions from the burn characteristics, they also increase the demands on ignition. In many cases, this increase exceeds the capabilities of the OEM fixed spark ignition system and reduced mileage and performance result from misfires.
   Two approaches are presently used to achieve the advantages of lean air/fuel ratios (low emissions and potentially good fuel economy). One is the stratified charge (Honda CVCC approach; the other is the aftermarket computer ignition method. The stratified charge approach requires a specially designed engine with an ignition and combustion chamber joined by a small hole. The sparkplug is in the ignition chamber, which is small. In this chamber the fuel system is set up to maintain about an 11: 1 air/fuel ratio, ideal for easy ignition. In the combustion chamber, the fuel system maintains ratios as lean as 25:1.
   Such a ratio could not be reliably ignited by most electric sparks, but once ignited it has very efficient burning properties. When the spark occurs in the rich ignition chamber, it should easily start the small amount of fuel burning under pressure. This causes a jet of flaming gasoline to spray out from the ignition chamber through the hole into the large combustion chamber, where it ignites the fuel. Remember that an air/fuel ratio too lean to be ignited by a spark can still be ignited by a flame.
   
   
4. How does an electric spark ignite fuel? Actually, nobody is quite sure. Three long-standing theories exist. The first, the thermal theory, has been around the longest but has recently lost ground to the other two, as new, more sophisticated tests are run. The thermal theory states that ignition is simply a factor of the heat of the spark, and that it occurs in much the same way as if the air/fuel mixture were lit with a burning match. This theory has been somewhat (although not entirely) discounted by recent data compiled on electronic ignitions, which have the ability to tailor a spark's duration (burn time), intensity (strength, measured in amps) and phase angle to dictate when during the spark duration the highest intensity will be delivered. The ability to control ignition through factors other than thermal energy has led to the metal fragmentation and ping pong theories.
   The metal fragmentation theory states that highly ionized electrons rip metal fragments from the sparkplug's electrodes as they migrate across the plug gap, collide with the hydrocarbon molecule (gasoline), and act as a catalyst to the burning process. Supporters of this theory allude to tests that show how variations in the electrode's metal composition will enhance certain ignition properties. As a practical example, Champion sparkplug electrodes with a relatively soft metal composition allow for large amounts of metal fragmentation, which enhance the plug's ability to burn through large quantities of fouling contaminants. This formula is useful for engines that have a fouling problem, although the plug's soft electrodes tend to wear down relatively quickly. At the opposite end of the scale are Autolite plugs, which feature a hardened metal composition electrode with long-wear characteristics for engines that don't have a serious fouling problem.
   The ping pong theory states that fuel ignition is accomplished when electrons from the spark collide with hydrocarbon molecules as they migrate across the electrode gap, splitting the positively charged protons from the molecule, which in turn collide with other molecules, triggering ignition. Although fuel ignition may well be a combination of all three theories, recent trends lean toward a combination of the ping pong and metal fragmentation concepts.
   
   
5. Ignition timing for modified vs stock engines. It is worthwhile to understand what ideal timing is. If you had an ideal ignition, the piston would come up to the top of the compression stroke, then ignite all the fuel simultaneously so the burning push would all be downward. At Jacobs, we are presently working on a microwave ignition which replaces sparkplugs and wires with wave guides. The combustion chamber becomes a little microwave oven. Regardless of rpm or engine volume, when the piston is 4 degrees before top dead center, a blast of microwave energy is directed down the wave guide into the combustion chamber. (The 4 degree advance maintains the connecting rod and bearings under pressure (compression) rather than allowing the stress to reverse and have them go into tension.)
   Within millionths of a second this energy rattles around to every corner of the combustion chamber, and heats all the gasoline to well over its ignition temperature. Bam-it all goes off at once, and since the timing is so retarded, there is no detonation. The entire flamefront push is down rather than having the flamefront partially oppose piston travel, as happens with conventional sparkplug ignition and normal timing curves. Also, air/fuel ratio is no longer important since you don't count on traveling flame propagation.
   Unfortunately, at present, no available ignition can ignite all the fuel simultaneously and there is a fixed amount of time lost as the fuel starts to burn. To compensate for this lost time, timing advance is added, and since the engine turns through more degrees at higher speeds during this fixed time, more advance is needed at higher rpm.
   The burning rates of fuels vary markedly. Gasoline in a high compression modified engine starts burning very fast, then slows down much like a high speed, lightweight rifle bullet. In a low compression "smog engine," the other extreme, it burns slowly at first, then, mostly due to combustion chamber design, EGR and low octane unleaded gas, accelerates much like a rocket, and for this reason, smoggers like a lot of advance at low engine speeds, but less total advance at the higher speeds. This means initial timing should be about 16 to 24 degrees on a smogger but the total centrifugal, plus initial timing, shouldn't go much beyond 30 degrees or that last rush of burning acceleration will eat the valves and pistons in about 15,000 to 18,000 miles.
   High compression street engines, on the other hand, want about 6 to 8 degrees initial timing to allow starting, but ran much more responsively with more initial advance, say 16 to 22 degrees. The total (initial plus centrifugal) timing of 38 to 45 degrees they can tolerate results from slower burning at the end of the cycle.
   The problem is that fuel starts to burn the moment ignition occurs, and during the entire advance time, burning is fighting piston travel and wasting energy. The balance of down-pressure is higher with the proper advance. If the advance is too much, you lose power and mileage. To make matters more complicated, the timing light does not tell you when ignition occurs; what it does show you is when the leading edge of the high voltage spark occurs.
   If you wish to know the best timing curves for your particular engine/vehicle combination, spend two hours, and using the techniques in the timing section of Chapter 5, dial in your absolute optimum curve.
   What's the difference? Let's see what happens when a spark, which is 1.5   milliseconds (.0015 sec.), needs 1 millisecond (.001 sec.) to get combustion well under way. In 1 millisecond the engine rotates through 18 degrees at 3,000 rpm. Your timing light may show, say, 36 degrees advance, but appreciable burning is really progressing only after 18 degrees advance. This retarded timing leads to tremendous loss in fuel efficiency. The engineer is caught between a rock and something hard because if this lag did not exist, timing would need to be set at 18 degrees. Otherwise peak pressure would occur before TDC, and the engine would be fighting itself due to over-advance.
   The time between when the ignition voltage begins and the mixture starts to burn is called ignition lag, and is a serious problem with today's mixtures. It is much more of a problem on alternate fuels because they start burning very slowly, and the plug gap must be made smaller.
   If you could count on ignition lag to stay constant, it would be easily compensated for by additional timing advance. The main problem is that the lag is not very predictable, causing a shift in the time of peak cylinder pressure with respect to piston position. The effects are the same as if you had severe spark scatter. However, in ignition lag the timing shift moves in waves rather than within each cylinder. In most engines, lab measurements of cylinder pressure vs piston position or crank angle show a scatter much greater than any normal spark scatter. Any ignition or engine change that can reduce this scatter can potentially increase power and efficiency by making it possible to uniformly set timing to attain peak cylinder pressure at the optimum crank angle more of the time.
   Normally any reduction of ignition lag will be accompanied by a corresponding reduction in ignition lag scatter, leading to improved performance.
   Ignition lag can be divided into three parts. The first, called voltage rise time lag, is the time lapse from when the spark voltage first starts and your ignition timing light flashes till spark current begins. (Remember, it's current, not voltage, that causes ignition.)
   The second, called transport lag, is the time lapse from when the spark current starts until a ball of flame is initiated. This ball of flame starts at the diameter of the sparkplug gap.
   The third part of ignition lag is the time it takes for the diameter of the ball of flame to grow from the sparkplug gap to approximately 0. 100 in. This is called growth lag. After 0.100 in., the flame propagates very rapidly.
   There are only two things you can do to reduce ignition lag. The first is to improve the spark profile, which deals with how intense or long the spark is, not with timing per se. Because most engine modifications in general, and off-road use in particular, favor a different spark profile than stock engines, the original equipment ignition used on modified power plants has a very long transport lag, particularly under cruise conditions.
   The second is to reduce growth lag by increasing the sparkplug gap nearer the magic 0.100. This has the added advantage of reducing transport lag because it kicks off burning in a shorter time. While not understood theoretically, most of us working in this area have found that further improvements can be had by increasing plug gap on extended tip plugs as opposed to recessed tip. Unfortunately, when going to higher compression and rpm, you often have to reduce spark gap to meet stock ignition limitations (remember it was not designed to ignite under these conditions).
   Here again, some of the exotic ignitions discussed in Chapter 3, as well as a computer ignition, have two advantages. First, they continuously change the spark profile for each cylinder as it generates each spark. As the fuel is changed the unit completely re-tailors the spark profile to accommodate the new fuel condition. Second, with the higher, controlled output, you can actually increase sparkplug gap even with higher compression modified engines. All common ignition systems do some of this inherently within the maximum output limits of the system. The actual output voltage only reaches the voltage necessary to fire the plug under the given conditions. Higher cylinder pressure and leaner mixtures require higher voltage to fire a given plug gap. After the plug fires, voltage across the plug drops from the 10 to 40 KV necessary to a much lower voltage, I to 4 KV. This lower-sustaining voltage depends on the mixture in the gap, pressure, turbulence, and spark current. Spark current is generally limited by characteristics of the specific ignition system. Inductive ignitions, such as OEM and magnetos, generally supply spark currents in the .020 to .060 amp range; CD ignitions in the 0.100 to 1 amp range.
   Unfortunately, most ignition changes are detrimental to engine operation when they are brought about consequently from changes in combustion chamber conditions. For example, just when you need the most ignition voltage, a fouled sparkplug causes an actual reduction in available spark voltage, as does more combustion chamber turbulence cold starts with a weak battery. Another example is that due to impedance matching, on a hot dry day, under warmed-up cruise conditions, spark energy inherently increases just when the engine needs less energy. In both these examples, the ignition system is inherently adapting to engine environmental and running conditions. The problem is that IT IS CONSEQUENTLY ADAPTING SPARK CHARACTERISTICS IN THE WRONG DIRECTIONS SUCH THAT THE ENGINE WILL NOT BE RUNNING AS WELL.
   On the other hand, there are ignitions that actively control spark characteristics to favor improved engine efficiency. For example, they increase spark voltage in the face of a fouled plug. This improved spark profile, which would also allow for increased sparkplug gaps, leads to approximately 21.9% better fuel efficiency on modified engines and 16.3% on stockers. Power is also improved. Any ignition that will adopt spark in response to engine needs puts the zing back in, particularly at mid-range and higher rpm. Engine dynamometer tests show a reduction of 74% in ignition lag between a spark with good profile and a weak one.
   Ignition lag in this context only refers to the time it takes for the flame ball to reach 0.100 in., so by reducing ignition lag 74 to 75%, you couldn't reduce your timing from 36 to 9 degrees. You could only make variations on the order of 6 to 12 degrees because the burning process, which is not under the control of the ignition, also takes a finite time. In most cases, you wouldn't reduce your timing very much because, except for use with racing fuels, timing is set such that there won't be pinging or detonation. However, as it is now, pinging is usually caused by the CYLINDER WITH THE LEAST IGNITION LAG, because any given cylinder would react the same when the timing is set at 45 degrees with 12 degrees of ignition lag as it would if the timing were set at 35 degrees with 2 degrees of ignition lag, or at 33 degrees with a theoretical 0 degrees of lag. It only detonates once burning really gets underway, not when the spark starts.
   While conditions vary from engine to engine, and even between cylinders on the same engine, it is not uncommon for ignition lag to vary from 2 to 12 degrees depending on many factors, from temperature to fuel quality, and of course, driver technique. For example, a cylinder, under immediately existing conditions, would have an optimum timing of 33 degrees advance. In fact, it would start to detonate with much more advance. Under general conditions, where ignition lag varies from 2 to 12 degrees, you would be forced to set your timing at 35 degrees to avoid detonation because under those extreme cases, ignition lag could be only 2 degrees. For the same cylinder, 33 degrees optimum timing, when the immediate existing conditions changed to a 12 degree ignition lag and you had set your timing at 35 degrees, the cylinder would be working with an effective timing of 23 degrees. This is 10 degrees retard from optimum, leading to all the resultant power and mileage loss. You obviously can't set your timing at 45 degrees because when ignition lag dropped below 12 degrees, you'd be into engine-damaging detonation.
   Certain techniques are used, such as vacuum advances and electronic-controlled or mechanical canisters, but they only work in a general way and do not meet the specific needs of each cylinder on each stroke of the engine. If, on the other hand, by actively controlling ignition spark properties you could reduce the range of ignition lag to a maximum of 2 degrees and a minimum of 1 degree, and reset your timing to 34 degrees (only a 1 degree retard), you would always be operating at or near engine optimum. You would experience the effects of about a 7 degree average timing advance with increased mileage and performance even though you had actually retarded your timing. This is why there is so much interest now that the technology has advanced to the point where we can actively control full spark profile.
   
   
6. Spark quality: very different requirements once you've modified the engine. It is important at this point to understand the difference between spark timing and spark profile. Timing controls when spark is delivered and burning starts. Spark profile determines what is delivered. Both are equally important; but unfortunately, timing, being easier to understand, receives the most attention.
   The interrelation between the two is like sending a birthday present. You have to send it earlier than the birthday (timing advance) to allow for travel time, and it has to be the appropriate gift (spark profile) to fill the recipient's needs/wants (ignite the fuel).
   Just as timing must vary to reflect varying shipping conditions, so must spark profile be varied to reflect the recipient's changing tastes (engine requirements). Unfortunately, ignition systems without active control put out sparks which are passively controlled, leading to greater ignition lag.
   To reduce the percentage of misfires, engineers have usually taken what is called the classic approach. This consists of trying to estimate how the engine will be used; that is, what percentage of the time it will be hot vs cold, accelerating/climbing vs cruising, what vehicles they want to put that engine in, the type of fuel expected over the life of the engine (this is a tough one, as we are presently finding out). Then they design an ignition system to put out a spark that is right in the "middle" of the expected requirements and, using this classic approach, they can get good reliable ignition 88 to 94% of the time. Unfortunately, 6 to 12% of the time, while the ignition is putting out good, healthy-looking sparks, the real life street requirements are so far from the engineer's expected "middle" that the spark cannot kick off the fuel burning for that cycle. However, because of slightly different conditions, the same spark may well get combustion on the firing cycle. This classic approach is to write off that last misfire in order to play the overall odds.
   A "middle" spark requirement for a typical pure factory engine goes right out the window when engine modifications are made. The 88 to 94% ignition, which the engineers were so proud of, can drop to about 71-77% with modifications.
   How far off this "middle" spark the modifications move the engine's requirements determine the extent of the loss. (Did you ever wonder why some modifications give so much better results than others? The answer usually has to do with how far of the "middle" spark each modification moves the engine's spark requirements.) There are steps that can be taken that lead to astounding improvements in a modified engine's mileage and vehicle performance. But first we must understand what to look for and what conditions affect spark requirements. Unfortunately, almost everything does. For example, engine temperature, altitude, quality and composition of the fuel, air/fuel ratio (which varies markedly from cylinder to cylinder), driving habits and techniques (throttle pressure), even cylinder-to -cylinder variation in compression, all affect what kind of spark an ignition system should ideally generate, even the degree to which fuel is slopping back and forth. (Did you ever have something test out really well on a dynamometer and not work at all in actual use? Did you ever wonder why EPA mileage figures are so different from your experience? Many EPA-type emission and mileage tests on new cars establish baseline data to evaluate the effects of future changes. These have been run by myself and many others in the private sector. We rarely find any new cars that come even close to the EPA mileage claimed. I have concluded that EPA mileage tests must have been done on highly tuned pre-production samples and are not really representative of production cars. Another common reason for the mileage difference is that fuel is relatively level on a dyno but vigorously sloshes back and forth running on the road and especially when running off-road. I've seen as much as a 15% difference on the same vehicle.
   For maximum fuel economy, you had better have your probability of misfires very low because at 55 mph it takes between 120 and 200 ignition sparks per second just to keep you rolling. It does not take a very high percentage of misfires before you have dumped a lot of unburned fuel into the exhaust pipe where it will probably ignite from the heat of the exhaust manifold. When this happens, no propulsion energy has been achieved and pure waste has occurred. Fortunately, 35 mph is the optimum speed range for OEM ignitions and if everything else were theoretically perfect, misfires could be below 1%. Unfortunately, we don't drive at a steady 35 mph all the time and rarely if ever is the fuel or fuel system perfect.
   A common lament is, "I did everything for better mileage/power. I installed a better carburetor, mixer, headers, cam, and the mileage/power went down' " What has happened is that to improve burning properties, the owner substantially moved his "middle" ignition requirements. What he gained from better burning he more than lost from increased misfires. The lesson is quite clear: any time you make a change that will affect combustion chamber conditions, you had better make corresponding changes in the type of ignition spark delivered or you will be playing Russian roulette with your fuel efficiency.
   This is especially true with the modified vehicle. Note that we are not talking about timing changes, but rather the composition of the actual spark, what the engineers call spark profile. Even though spark profile sounds complicated, there are really only three things that completely describe it They are duration, that is, how long the spark goes on; intensity, that is, how much current is being pumped through the spark gap; and phase angle, which describes when during the spark you pump most of the current through the gap. When dealing with phase angles, you ask a question like, "Do you want the current to start small and build rapidly as the spark goes on?"-ideal for starting and cold engine operation. The ideal phase angle for cruising, particularly with lean mixtures, is to have low level, even current for the full length of the spark. The ideal phase angle for stock engines is wrong for modified engine use.
   If you have modified your engine and would like an estimate of how you should appropriately modify your ignition, you can call our technical staff at (800) 6278800. Tell them what engine you have, what you want to do with it-race, street, tow- and what modifications you've made!
   To correct for phase angles which are off from optimum, many knowledgeable testers came to the conclusion that higher spark energy improves performance in the real world, even though they found that by simply increasing energy, the improvement was slight. For example, many ultra-high energy systems, like plasma jet, are judged not to be worth the added cost and weight because improvement is minimal. Microwave ignitions, which can ignite all the fuel in 1 or 2 degrees of engine rotation right near TDC, do produce exciting results because piston pressure caused by fuel burning is virtually never fighting piston travel. That is, the pressure is always pushing in the direction of piston travel. Here again, cost and safety more than weight is the limiting factor.
   Even the experts, who go for the more-spark-energy-is-always-better theory, realize that once given enough energy, you will get cavitation or punch-through. Normal automotive ignitions, when acting property, operate in a range called glow discharge region. After the initial spike initiates current flow, this region of operation is characterized by a nearly constant voltage across the plug gap over a fairly wide range of spark current, 0.1 to approximately 1 Amp. In this region of operation the voltage increases with an increase of spark length (spark gap) or in gas pressure (compression). Punch-through or cavitation could be said to occur when the spark current is increased to the arc discharge region of gaseous discharges. When this happens, an abrupt and dramatic decrease in spark gap resistance occurs, causing a correspondingly sudden decrease in gap voltage, from over 1,000 to under 100 volts. Since spark current is usually limited by the internal impedance of the ignition system, this abrupt decrease of gap power occurs, usually in excess of 10 to 1! This is because electrical power is equal to volts times amps and if the volts decrease by a factor of ten across the gap, even though the amps increase slightly, there is a dramatic decrease in effective ignition power. This phenomenon does not normally occur with conventional inductive or even CD-type automotive ignitions because their spark currents are normally well down in the glow discharge range.
   The word normally means most of the time. However, as discussed earlier, everything in ignition is probabilistic, meaning it moves all over the map. You can get cavitating on this cylinder this time and not on the next, or even again on this cylinder next time. Remember, "It's really not that simple". It isn't even so simple that you can say cavitating always leads to misfires. It doesn't. It simply significantly increases the chance of misfires just as ignition systems that categorically increase spark power past the optimum average significantly increase the chance of cavitating ("It isn't that simple!").
   What is that simple is that every time cavitating leads to a misfire, it always robs power, performance and mileage. So, if you get 6% misfires from cavitating, you'll get a corresponding loss in power and mileage.
   For any given ignition system combustion chamber conditions can have a profound effect on whether cavitating occurs. Conditions that contribute to its likelihood are low turbulence and small sparkplug gaps. Both these conditions are likely to be present when retracted gap racing plugs are used. Here again, any modification of factory specifications moves combustion chamber conditions away from the expected middle ground. This leads to increased misfires.
   Another factor which seems to contribute to this problem is the use of nitromethane fuel, which is a polar solvent and therefore good electrical conductor (spark short-circuiter). Extended tip plugs with large gaps are more efficient than retracted gap racing plugs at igniting fuel, but when you want lots of power, how do you make an extended tip plug that will not overheat with resulting pre-ignition and even detonation in, say, a blown nitro motor?
   In general, the optimum spark profile varies as follows:
   * Low rpm low load- Long time duration low current spark can be effective due to long relative time available to establish flamefront.
   This long duration-type spark can also aid in establishing a flame kernel by
   adding to the heat of combustion during the critical time between spark initiation and the time the flamefront becomes self-sustaining under these conditions.
   * At high rpm heavy load a shorter duration but higher current spark is optimum due to the shorter time available to establish a flamefront, along with much bigger turbulence and charge density. Under these conditions it is often possible to increase power output somewhat by means of a CD ignition due to its much higher spark current, even though its spark duration is generally shorter, and this short duration can get you into trouble at low speeds with light throttle. Some CD ignitions attempt to get the best of both worlds by using multiple sparks at low rpm.
   Since spark energies and combustion chamber conditions vary independently and usually in a way that spark energy and voltage decrease when needed most, it came to pass that engineers thought (or maybe "felt" is a better word) it was a good idea to give a real hot spark all the time. Present testing says this is wrong and in fact, can lead to cavitating.
   For those of you who aren't into electrons, glow regions and other molecular phenomena, and want to know what cavitating is in layman's terms, it's simply this: just as too much torque can cavitate a boat prop, so cavitating is where the spark punches a "hole" in the air/fuel mixture. After pumping a lot of current through this hole, the spark ends and the air/fuel flows back over the hole without igniting (a misfire). Cavitating is very common with gasoline and modifications, such as competition cams, and so forth, especially those requiring reduced sparkplug gaps. This is especially true at light throttle and mid-range engine speeds.
   Well, if it's that simple, why don't engineers back down on the spark energy? Because with present factory engine design, and especially with modified systems, you need every bit of that energy, and then some, to get today's gasoline engines started and keep them running smoothly when they are cold. This, then, is the inherent problem of the classic fixed spark system, both for contact point and the newer electronic type ignitions.
   There is a proven superior approach. Rather than crudely trying to compensate for engine environment, running conditions and cylinder-to-cylinder variations, such as BMW, Mercedes, and Datsun try to do by tweaking fuel injections, carburetor, accelerator pumps, and so forth, why not compensate the individual spark as it is being generated cylinder by cylinder so it will always be appropriate? This is commonly called the "adaptive-spark" method. To achieve this adaptive spark it was necessary to design and develop a high speed sensing computer ignition.
   Prior to the development of this computer ignition, there were two major unsolved questions related to the "adaptive spark" method. First, how do you ask an engine what it wants? Second, how do you do anything about it fast enough to improve fuel efficiency; that is, before the spark is over? This is a particularly difficult problem with modified vehicles because of their tremendous variation in spark requirements. The answer to the first question is to have the computer measure the changing electrical resistance of the actual spark as it is sparking through the combustion chamber. Everything that affects ideal ignition and burning also affects the electrical resistance of current passing through this same fuel/air mixture. Each engine has its own distinctive resistance pattern. The change in resistance instantly feeds back to the ignition wires, through the rotor, and back through the coil. Notice that this makes each sparkplug an "in-cylinder sensor" for its own combustion chamber.
   This changing spark resistance is the same information the mechanic reads from an engine oscilloscope, but in the case of a computer ignition, the information is fed into the high speed, on-board analog computer, which quickly picks up and uses the information to change the spark as it is being generated. This change in spark thereby causes another change in electrical resistance which the computer quickly picks up and uses to make another change in the spark until, by progressively "seeking" toward the optimum profile, it inevitably finds the right spark (less the 0.0009% inherent misfires) to start any (gasoline, propane, CNG, etc.) fuel burning.
   Whenever any significant ignition improvement is made, such as changing from an inductive ignition or magneto to a high energy CD, or especially a computer ignition, some adjustments of timing and fuel system will probably help you take full advantage of the improved ignition:
   * Increased spark current should reduce ignition lag, effectively advancing ignition timing. Some retard may prove beneficial at high rpm.
   * Some custom tuning of the fuel mixture will be beneficial because the original mixture was probably limited by ignition characteristics.
   * Improved ignition tends to broaden and make less critical most engine tuning so you can try for slightly more radical adjustments in general.
   When our labs tested fixed (as opposed to adapted) spark systems we found that a combination that worked well on one vehicle actually caused deterioration in mileage and performance on another, or on another fuel. This in itself was not surprising, but since it only takes a small variation in production tolerances to make a big difference, there were two cases where a combination that improved a particular model actually caused deterioration in the mileage and performance on another vehicle of exactly the same engine!
   
   
7. Proper sparkplug heat range. One important ignition consideration is the heat range of the sparkplugs. This is particularly a problem with the dual-fuel conversions. Heat range refers to how fast combustion heat is drawn away from the sparkplug center electrode. The center electrode is right in the middle of combustion, and therefore absorbs heat. A center electrode which is short and squat is cold, because heat has a short travel path to the water jacket. A long slender center electrode is hot because it has a long path to the cooling water. If the plug is too hot, it fractures from intense heat and/or detonation, which this hot tip causes.
   The disadvantage of a cold sparkplug is that carbon deposits form on the sparkplug tip and this quickly leads to fouling. it's best to use a plug which is not so cold as to foul but not so hot as to destroy itself or the engine. While this is difficult enough on any stock engine because of the tremendous difference in engine environment and running conditions, it's virtually impossible on modified engines. For this, you want a colder plug on a modified engine, but then the cold plugs have an even greater tendency to foul because these same plugs are far too cold, and tend to foul in a way that, while not obvious in driving, cost a lot in fuel efficiency.
   There are two possible solutions. One is to change plugs when you change driving conditions. The other solution again relies on a class of devices which have the ability to sense combustion chamber conditions, then optimize the spark to meet the sensed conditions. When they sense carbon is depositing (fouling) on the plug, they increase the intensity of the spark drive to accomplish two effects: first, maintain the proper spark profile even though the fouling carbon is robbing spark energy; and second, automatically run a high intensity stream of electrons down the carbon in order to vaporize (ionize) it. You cannot burn carbon off a sparkplug gap. You have to use the intensity of this stream of electrons to chip the carbon away.
   Another solution is a capacitor-discharge ignition which can help arc such plugs. CDs generate such large current that the voltage through the carbon is greater than voltage to spark the gap. However, uncontrolled capacitive discharge systems introduce other ignition problems, such as very short duration, cavitation, and energy loss.
   Energy of any kind, including spark energy, is power multiplied by the duration during which that power is supplied:
   ENERGY = POWER X DURATION
   It follows that for a short duration spark to achieve a reasonable supply of energy it must supply a lot of power during the short time it is arcing. For any given sparkplug condition, once arc-over is achieved, spark power is roughly proportional to spark current. Therefore, to generate sufficient spark energy, capacitor discharge ignitions usually generate much higher spark currents than other ignitions. This is one way they achieve their resistance to plug fouling.
   On the other hand, losses in the coil and secondary wiring resistances also go up as spark current goes up, which can result in less total energy actually reaching the spark gap, even though the energy generated was the same. Resistive power losses increase as the current squared; therefore, power losses increase by a factor of 4 for each factor of 2 increase in spark current. Since most CD ignitions have a spark current 4 times as high, but only 1/4 as long as OEM inductive ignitions, potentially 4 times as much energy resistive losses can occur. In equation form:
   Spark Energy Losses Due to Resistance in Series with Spark Current
   
   E._loss= P._loss x T
   P._loss= I²R
   E._loss= I²R x T
   
   For a CD System with Current = 4 and Duration = 1
   E._loss(CD)= 4² x R x I = 16R
   
   For an Inductive System with Current = 1 and Duration = 4
   E._loss(Ind) = I² x R x 4 = 4R
   
   Loss Ratio of CD to Inductive
   E._ratio= 16R/4R = 4:1 where
   E._loss = general loss in spark energy due to series resistance
   R = value of series resistance
   T = time during which spark current is flowing
   E._loss(CD) = loss in spark energy when using a CD
   E._loss(Ind)= loss in spark energy using an inductive system
   E._ratio = ratio of energy losses CD to inductive
   
   Unless some steps are taken to reduce secondary resistance, losses may become excessive. The following are sources of resistance:
   * Sparkplug wires should have less than 5,000 ohm resistance. Solid wires are lowest in resistance but can cause radio interference and crossfire between wires unless care is taken in wire routing. Special wound wire core sparkplug wires are available which are a good compromise for most uses.
   * Resistor sparkplugs cause additional losses and normally should not be used with CD systems in full race applications.
   * On the other hand, increased sparkplug gap improves effectiveness of CD systems if secondary voltage capability is adequate, which generally means you should have 8mm secondary wiring and large diameter distributor cap. Also, many CD ignitions do not adequately compensate for reduced battery voltage during cranking. When using one of these ignitions, maximum usable sparkplug gap may be limited by starting considerations. Also, whenever using larger-than-standard sparkplug gaps the whole secondary system must be kept in tiptop condition, which means clean and dry. Note some CDs use tricks to extend spark duration or multiple -discharge techniques to improve low rpm operation. In this way, both computer and CD type ignitions allow you to use the really colder plugs, which are ideal for the modified engine, but not have them foul when you drive around town.
   
   
8. Looking for the perfect 10 in ignition coils. The perfect ignition coil has not yet been invented, but there are some very good 7's and 8's, and even one 9. The problem is that an ignition coil is inherently a mass of conflicting push-pull design objectives so you can never reach a perfect 10! What would make a coil good in one area, say, quick starts, would cost you in another area, like high rpm capability. It's all a compromise, but some coils trade off considerably better than others.
   Let's look at the design maze of objectives you are trying to achieve.
   * High energy spark for starting
   * Not too much spark energy when running, or you have sparkplugs which run too hot, causing them to wear out. Also, excessively high running spark energy leads to power-robbing ping, knock and cavitation. The street symptoms of ignition cavitation are: detonation, surging, and poor gasoline mileage.
   * An ever-present problem is that battery voltage works in direct opposition to the desired spark energy. When you are starting, you want the highest spark energy when battery voltage is lowest. When you are warmed up and running, you want lower spark energy, but now the battery is at its highest. Even though the ballast resistor and current limiting circuitry partially compensate for this conflict, help is only minimal.
   * You want a lot of turns on the secondary side for high spark output voltage, but the more secondary turns, the lower the spark current, which can cause reduced high rpm spark energy. (Spark current approximately equals primary coil current divided by coil turns ratio, where turns ratio is the number of secondary turns divided by the number of primary turns.)
   * You want fast rise time (the time it takes for the voltage to reach the needed arc-over level), but a spark that rises fast usually ends fast, and you also want enough duration to assure good reliable ignition.
   * It takes energy in to get energy out. For a coil operated in the inductive mode, energy in means primary current from the battery. If you reduce the current in, you reduce the spark energy. On the other hand, large primary current means quicker battery rundown and higher strain on electronic ignition amplifiers or on contact points.
   * You want a core with very high permeability and saturation for high energy storage. That means the higher the spark, the more energy stored in the core, but the longer it takes for the battery to build up core energy. If they're too high, while you would have a terrific low rpm spark, you wouldn't be able to go any faster than 30 miles per hour. One way racing or high performance coils get higher rpm capability is by keeping permeability and saturation low for the quick recovery needed at high rpm, but they can pay the price of lower energy (which reduces the ability to fire fouled plugs and the capability to run well on the street, especially when it's cold and damp). The design philosophy in high-performance race coils is, "The owner will be changing his plugs often since it's a racing application." Race coil designers won't worry about around-town driveability.
   
   
   With all these conflicting design objectives, how can one coil be better than another in almost every area? There are, of course, some obvious ways, such as using bigger gauge wire (even though the number of turns remains the same), using purer alloys, or insulating materials. To understand beyond this it is necessary to know how a coil works in the first place. Fortunately, they all work on the same basic principles, so once you learn them, you will be in a good position to make judgments and even run tests on any and all coils.
   The spark-generating part of the coil contains only three functional elements:
   * A primary winding, which is powered by the current from the car battery.
   * A secondary winding, which usually has about 100 times the turns of the primary.
   * The core, which is made of a passive magnetic metal, except for certain types of variable magnetic coils, which are made of active variable metal. Both the primary and secondary windings are wrapped around the core which magnetically "couples" the primary winding to the secondary.
   In the d-c mode, a coil works by allowing current, supplied by the battery, to slowly build up in the primary of the coil, making an ever-stronger magnet of the core metal. How strong a magnet for each bit of current is called the permeability of the core. Air and other non-magnetic materials have very low permeability. That is, a large amount of current produces very little magnetism. Since the change in magnetism is directly related (double the change and you double the output) to spark generation, coils with higher permeability have more output. The amount of magnetism is also proportional to the amount of current, so increasing current increases magnetism, and perhaps more important, decreasing current decreases magnetism.
   This would all point to having the highest permeability for the highest output. Unfortunately, high permeability has its trade-offs. (Remember, it Is "not that simple.") High permeability means the core magnetizes slowly so rpm is limited. Also, as an unfortunate inherent characteristic, the higher the permeability, the lower the amount of total magnetism the core can hold.
   If you keep building current in the primary, eventually you will reach a point where the core is so full or "saturated" that more current does not produce any stronger magnet. Everything (except government spending) has a limit. It is a basic law of nature (physics) that this process of having primary current make a magnet out of the core takes time. It is the electrical equivalent of getting a heavy object moving. The pushing force is the battery voltage, the weight of the object is the permeability, and saturation is how fast you can ultimately get it going. High permeability and saturation mean you need a lot of time for the entire process to take place.
   There is another law of nature needed to make coil operation possible. To
   understand coils you must have a feel for this basic principle: If you think of magnetism as a light ray, and you put up loops of wire such that the light shines through these loops, then voltage will be generated in proportion to the number of loops times how fast the light (magnetism) intensity is changing (not how much light, but how fast it changes: a lot of steady light produces no voltage).
   Using this close visible analogy, the core is a flashlight converting battery energy to light energy. The loops of wire generating the voltage are the windings on the secondary (coil output). Now we must look at the rate of change of the magnetism (light). Remember, the current builds slowly in the primary because the process of converting current into magnetism takes time. Therefore, during current build-up, the rate of change o the magnetism is slow, as is output voltage. During this slow current build-up coil output voltage is nowhere near strong enough to produce a spark.
   The illustration shows current building up by passing through some kind of switch, such as points (older vehicles) or an electronic ignition (newer vehicles) (Even the most modern electronic ignitions, like the General Motors unitized, simply substitute an electronic switch fo the old contact points.) Even though it took a long time for the battery to buil up coil primary current, obviously, if you open the switch, you stop the current very quickly. After all, where can it go if the switch is open? Since magnetism proportional to current, you change magnetism from the highest to zero very quickly, and this quick-change produces a BIG spark voltage! (Remember voltage is proportional to how fast the magnetism is changing, not how much there is. Nature doesn't care if the change is positive or negative.) Even though a negative change reverses the polarity of voltage, magnetism remains the same.
   Using our analogy of getting the heavy weight moving by applying steady force this opening of the switch is equivalent to slamming the moving weight into a wall. So violent is the current interruption th a condenser (equivalent to a pillow) needed to help absorb some of the shock or the wall (switch) would not be able to stop the moving weight and the wall would be knocked down (arc-over at the points, or transistor blown on electronics).
   Every time you want a spark you have to interrupt the current via opening the points or shutting off the electronic amplifier. For any given coil, the output spark voltage will be proportional to the amount of current (magnetism) flowing in the primary at the moment of interruption. Because it takes so long for current to build up (getting the heavy weight moving), the output spark voltage goes down as engine rpm goes up. Higher engine rpm means more frequent sparks which means less build-up time. (An eight cylinder engine at 2, 100 rpm, about 55 mph, must produce 140 sparks per second. A six cylinder at 55 mph, running at 2,600 rpm, needs 130 sparks per second. A four cylinder running at 3,000 rpm for 55 mph needs 100 sparks per second.) Once you reach sufficient rpm where the coil primary current is interrupted so frequently that the core doesn't have time to saturate, from that rpm on up, doubling engine rpm cuts build-up time and spark voltage in half. This is what gives you that very mushy gas pedal at higher rpm. It feels like the engine is starting to float or hold back.
   The one final law of nature which must be understood is:
   AS CURRENT (SPARK ENERGY) FLOWS FROM THE SECONDARY WINDINGS INTO THE SPARK, IT GENERATES NEGATIVE, OR CANCELLATION MAGNETISM. NEGATIVE MAGNETISM IS PROPORTIONAL TO THE NUMBER OF SECONDARY TURNS.
   It is current, not spark voltage, that ignites fuel! This fact is so misunderstood, let me state it again in slightly different words. Arc-over voltage is the voltage necessary to jump the plug gap. Any voltage more than that is bad because to get higher than arc-over voltage you need extra secondary turns, which means extra negative magnetism which in turn means less spark current, and again, IT IS SPARK CURRENT, NOT VOLTAGE, THAT IGNITES GASOLINE!
   Obviously, if you don't achieve arc-over voltage, no spark current will flow and you get a misfire. In a modern engine 12,000 to 24,000 volts will always
   be sufficient to produce arc-over. Why don't automotive electrical engineers design coils to just reach arc-over voltage? Three reasons: the first, discussed above, is that spark voltage goes down as engine speed goes up. If you designed the number of secondary turns to just achieve arc-over at 3,000 rpm, and the core was just saturating, then at 3,100 rpm you would lose engine power and waste gasoline mileage, but you would have an efficient spark up to 3,000 rpm. (In an automatic transmission the high end of passing gear demands the engine must develop power to approximately 5,000 rpm.) So most original equipment coils are designed to be adequate to about 5,100 rpm and are therefore forced to live with reduced cruising efficiency and smaller spark plug gaps during most cruising and street driving. (Remember, there are no 10's in ignition coils.)
   Second, required arc-over voltage is proportional to sparkplug gap. The bigger the gap, the more voltage you need to arc-over. But big gaps give a smoother running engine with better mileage and more power. For this reason, engine designers like big sparkplug gaps. (About a one-hundred -thousandth (.100") would be ideal.) Coil designers like small gaps because it allows them to use less secondary turns. If you'd like to know the best gap for your engine, write and tell me engine type, maximum rpm you will be running at, and what modifications you've made to your engine. We'll get back to you.
   Engine designers might win out if it weren't for reason three-spark plug fouling and wear. Sparks like to jump from sharp surfaces; conversely, sparks don't
   like to jump from round, smooth surfaces. When you put in a new plug, the center electrode is sharp where the flat cut borders the round side. After wearing awhile, the spark current, which always jumps from the sharp edge, eventually erodes it until it's round. This increases the required voltage by approximately 18%. In addition, this same erosion actually makes the gap larger (even more voltage is required).
   To make matters worse, sparkplugs also foul (carbon deposits on the center electrode partially or completely short out the spark energy) as they wear. Electrically speaking, a foul is the same as a grounded resistor in parallel with the spark gap. The worse the foul, the lower this energy-robbing resistance. A fouled plug has about one-half million ohms.) What happens as the coil starts to deliver voltage to the plug gap? This fouling resistor starts to conduct current, which is supplied by the windings of the secondary. With a fouled plug, each secondary winding subtracts magnetism from the core. Therefore, when and if arc-over is finally achieved, there is less spark current available to ignite gasoline. If the fouling resistance is low enough, it will rob so much secondary current, that is, generate so much negative magnetism, that arc-over will not occur (a misfire).
   With fouling, contrary to popular belief, you want to actually decrease the windings on the secondary so as to trade voltage for current. This is exactly opposite of what you want to do with a worn-out clean plug. Unfortunately, plugs usually wear and foul simultaneously, so engine designers are forced to reduce arc-over voltage requirements in the only way open to them- reduce plug gaps to way below the ideal of 0.100"! For several years, around 1975-76, General Motors tried to specify plug gaps up to .080" . Even General Motors can't violate the laws of physics. They were forced to reduce this specification considerably.
   There is one new approach in coils which might be given a 9 rating. Manufactured by several companies, it goes by various trade names, such as Auto-Output, Demand Output, Energy Coil, Ultra Coil, or Variable Magnetic. They use a new approach which can make them quite effective in inductive-type ignition systems. Since the number of turns on both the primary and secondary are fixed and permanent, and since the magnetism is determined by the number of turns and the permeability of the core, these special coils change their permeability automatically in response to the engine's needs! In effect, the core gives up more energy (magnetism) as the plugs foul and also as rpm increase. This holds the output voltage at arc-over under all engine conditions and yet develops ample spark current at all ranges of rpm, even with plugs which are worn and fouled (fouled to a point; obviously if they foul badly enough, lightning, or even a computer-controlled ignition, couldn't get a spark). Why not call variable magnetic coils a 10? There are two reasons. First, it doesn't let you open your plug gaps to the full 0.100 in., which would be ideal. However, it does let you increase the gap by about 30% over any other coil. This should lead to an 8 to 11% improvement in fuel efficiency. Second, we want to see what other new and inventive ideas come up in the future and hold the 10 spot open just in case.
   There are other less expensive than top-of-the-line variable magnetic aftermarket coils, of superior quality to many of the original equipment coils, such as Accel, Autotronics, Haynes, Mallory, and Nyome, and generally speaking, a good investment. However, don't buy a 50,000 volt coil unless you are prepared to do exclusively highway driving. (Your plugs will load up too quickly.) Don't buy an 8,000 rpm coil unless you really want to run 8,000 rpm. (You will pay the price in gasoline mileage when you are on the street.) Try to mount your coil in the factory-specified location. Automobile manufacturers already mapped out the right spot where you can be relatively sure it won't be excessively hot and have good air flow. To get rid of heat, the best coils are metal can fluid-filled and especially aluminum extrusion housed.
   In short, the secret is to buy a coil that matches your needs! If you do, your engine will run better and your mileage increase, which is what it's all about!
   A little off the subject, but one question I'm asked a lot these days of super-environmental sensitivitity is, "With the need to go to computer-controlled fuel systems, like Chrysler lean burn, Mercedes, Porsche, Datsun Zs, and General Motors C-4, and computer ignition systems like Energy Pak, Tri-Mag, Compu*Sensor or Control Corp., why don't we go to another means of energy besides the fossil fuels (gasoline, diesel, propane, CNG, etc.)?" The reason is in energy content. The best battery systems hold about 8,000 BTUs per cubic foot (BTU = British Thermal Unit-a measurement of energy). Some of the more exotic 100,000 rpm vacuum-housed flywheel cars (the present darlings of the EPA) can reach 64,000 BTUs per cubic foot, but pump gasoline, the kind you get at your regular service station, checks in at a remarkable 800,000 BTUs per cubic foot!
   The message is clear. Invest what you must to improve your present fossil fuel type engines! They will be your vehicle power source well into the foreseeable future.

More on...   

contact  info@castensdynojet.com for remarks on this website..
Copyright © 2000 Castens Dynojet Center S.L.
Last Update: September 15, 2000