Ignition and Combustion

March 25/98

Possibly nowhere else in the performance engine world is there more misinformation, myth and downright false claims than in the ignition component aftermarket. Many manufacturers of these products prey on the naive, uninformed consumer. We will try to help you understand what is really going on and what works and what doesn't. This article addresses current production performance atmo and turbocharged engines.

The Ignition/Combustion Process

Many people think that when the sparkplug fires, the fuel/air mixture explodes instantaneously, driving the piston down. If this really happened, engines would last only a few minutes before they literally grenaded.

Let's examine the process and dynamics involved from the moment that the intake valve is fully open. With the piston moving down the bore, cylinder volume increases, cylinder pressure decreases, allowing the higher pressure in the intake tract to push the fuel/air mixture into the cylinder. As the piston starts back up and the intake valve closes, cylinder volume decreases and cylinder pressure increases.

When the crankshaft reaches about 30 degrees before top dead center, the spark jumps the gap between the plug electrodes. The purpose of the spark is to raise the temperature of a very small portion of the fuel/air mixture above its ignition temperature. This is the point where true combustion begins. As the exothermic reaction starts, the mixture directly adjacent to the spark plug is also ignited and the process rapidly progresses out from the plug in a roughly spherical shape.

At about 20 degrees BTDC, the rate of heat release causes the cylinder pressure to rise above the compression line which is what the cylinder pressure would be at a given piston position without ignition. Notice that it has taken 10 degrees of crank rotation to generate this pressure level. This is known as the ignition-delay period.

The rate of pressure rise is a function of the rate of energy release vs. the rate of change of combustion space or cylinder volume. The rate of energy release is directly related to the flame propagation rate and the area of reacting surface. Flame speed is dependant on fuel/air ratio, charge density, charge homogeny, fuel characteristics, charge turbulence and reaction with inert gasses and the metal combustion chamber, cylinder walls and piston.

In technical terms, the pressure rise is referred to as flagregation. No two combustion cycles progress at the same rate or at a uniform rate. Some start slow and end slow. Some start slow and end fast. Some start fast and slow down. Generally, only the ones that end too fast will lead to knocking as the rapid pressure rise may happen too soon with the cylinder volume still decreasing or not increasing fast enough. Usually, not all cylinders will knock at the same time or on the same cycle because of this.

By the time the crank is at about 5 degrees ATDC, the cylinder pressure is about double that of the compression line. From this point to roughly 15 degrees ATDC the combustion process is very rapid due to the increasing area of inflamed mixture and the high rate of energy release. The peak cylinder pressure (PCP) occurs between 10 and 20 degrees ATDC on most engines and the combustion process is complete by 20 to 25 degrees ATDC.. The peak temperature within the combustion gasses will reach somewhere around 5000 degrees Fahrenheit and pressures may be anywhere from 300 to 2500psi depending on the engine.

Obviously it is very important to have the crankpin at an advantageous angle before maximum cylinder pressure is achieved in order that maximum force is applied through the piston and rod to the crankshaft. If the mixture was ignited too early, much of the force would simply try to compress the piston, rod and crank without performing any useful work. In a worst case scenario, the cylinder pressure would be rapidly rising before the piston reached TDC which would have the cylinder volume decreasing at the same time. This will often result in knock or detonation which is counterproductive to maximum power and engine life.

Detonation or knock is defined as a form of combustion which involves too rapid a rate of energy release producing excessive temperatures and pressures, adversely affecting the conversion of chemical energy into useful work. Detonation usually involves ignition and literal explosion of the end gases, these being the gases not in contact with the initial spark or the progressing flame front.

If PCP was achieved too late, again, less work would be performed. Most of the useful work is done in the first 100 degrees of crank rotation. Most combustion must be done with the piston in close proximity to the chamber so that the minimum amount of heat (energy) is lost into the water jackets and the maximum amount of energy is delivered to the crankshaft.

Let's examine the different variables regarding flame speed separately and their effects:

Fuel/air Ratio

Gasoline can be ignited in a non-stratified charge type engine between limits of roughly 10 to 1 (rich) and 20 to 1 (lean) air/fuel ratios. Most gasolines will burn fastest at ratios in the 12.5 to 13.5 to 1 range. The stoichiometric or chemically correct ratio is around 14.7 to 1 which also results in the lowest average emissions. Best power is obtained with a rich mixture of around 12 to 13.5 to 1.

Charge Density

Charge density is affected by the pressure and temperature of the charge. The higher the charge density, the more rapidly it will burn.

Charge Homogeny

This refers to how uniformly the air and fuel molecules are distributed in the charge. This is a very important factor with regards to successful ignition. If there is a big lump of fuel molecules with no air present between the spark plug electrodes or vice versa, at the time that the spark jumps, there will be no ignition. If the charge between the electrodes is leaner than 20 to 1 or richer than 11 to 1, there is little chance for ignition. Ideally, the molecules throughout the whole charge should be evenly spaced and distributed. This allows for a smooth rate of burn. If the charge is randomly mixed, there will be local variations in flame front propagation rates which will not produce maximum power as these may advance or delay when PCP is achieved. This phenomenon is known as ignition probability.

Charge Turbulence

Because the charge is in constant motion from the valve and port flow characteristics along with inertial effects and piston motion, the mixing of fuel and air molecules is dynamic. From one split second to the next, the actual mixture and molecular distribution changes. This can mean in some instances, that if a very short duration spark was initiated at one instant, the mixture might not ignite, whereas only a half millisecond later, conditions might be perfect for ignition. For this reason, very short duration sparks are undesirable. A long duration spark or multispark ignition system will ensure the highest ignition probability.

Fuel Characteristics

Low compression engines usually run well on low octane fuel because they have relatively low charge densities and the burn rate within these confines is usually predictable. A low compression engine switched to 118 octane race fuel will always lose power unless the ignition advance is increased to compensate for the slower burn rates. Even then, a low CR engine may lose power with the timing optimized for high octane fuel.

A high compression or turbocharged engine operates with much higher charge densities and consequently faster burn rates. The high octane fuel permits these rapid burn rates because it has far less tendency to autoignite and detonate under these conditions. As a result, high compression and turbo engines cannot realize their full hp potential without high octane fuel.

Inert Effects

Inert effects constitute 2 areas. Residual exhaust gasses left over from the last exhaust stroke tend to dilute the fresh charge and slow down burn rates. Camshaft timing, port flow and exhaust backpressure will affect charge dilution. Nitrogen is the major constituent of air and is essentially inert in the combustion process. Its presence substantially lowers the burn rate but since there is little that we can do about it, it is generally ignored. Nitrous oxide can be injected along with extra fuel to increase charge density as it contains a much higher concentration of oxygen than does air. Oxygen is the essential element in the combustion process.

The second inert effect concerns the relatively cold, metal engine parts in direct contact with the combustion gasses. Combustion will not easily take place in areas where the gas temperatures are well below the ignition temperature. This property is often used to advantage on engines to reduce the tendency to knock.

On many engines, a squish or quench area is used to negate combustion in certain areas to avoid knock. By having a matched area where the piston and combustion chamber come in close proximity at TDC, the gasses are kept cool enough so that they will not ignite until the piston has moved down the bore and cylinder volumes are increasing. This keeps the rate of pressure rise below the knock limit. Some people are dismayed when they install a thicker head gasket to lower the CR and have knocking worse than before. This is because they have negated the designed-in quench effect. A large squish area also tends to promote increased chamber turbulence which is important for mixing and power at high rpm.

Combustion Chamber Shape and Spark Plug Location

Combustion chambers and spark plug location and the number of plugs will have a marked effect on the time required to complete the combustion process. A large open chamber like a hemi which has a high surface to volume ratio, will combust more slowly than a wedge or modern pentroof chamber simply because it has more cold, metal molecules in contact with the combustion gasses which tends to slow reaction rates. For this reason, these chambers will require that the spark be initiated sooner to achieve PCP at the correct time.

The slowest combusting chamber would be an open chamber with a large bore size and the spark plug at one edge of the chamber. The flame front has a long distance to cover to complete combustion. By placing the plug in the center of the chamber, you halve the distance that the front needs to travel and will be able to reduce the spark advance needed to achieve maximum power. Another solution would be to add another spark plug to create two flame fronts which would also require much less time to combust. This is the solution in most aircraft engines where big bores and poor fuel distribution and homogeny require solutions to increase ignition probability.

Modern 4 valve engines with shallow pentroof chambers and a central plug location are fast, efficient combustors, requiring minimal advance for maximum power.

Inductive Discharge Coils

Generating the spark on most production automotive systems is accomplished by the coil. Coils have 2 sets of windings, a primary and a secondary. The typical coil will have around 250 turns of wire on the primary and about 25,000 on the secondary for a ratio of 100 to 1. The secondary section often uses an iron core to increase its inductance. Coil resistance on the primary will be from .5 to 2.5 ohms usually and on the secondary, between 5000 and 12,000 ohms.

The inductance and resistance of the coil will determine how quickly a coil can be charged and discharged.

A transistor is used to switch the current flow off and on in the primary coil. When the transistor is switched on, current rapidly builds from 0 to a maximum value determined by the coil inductance and resistance. This current flow induces a magnetic field within the primary. When the current is turned off, this magnetic field collapses which cuts the windings of the secondary coil and induces a high voltage surge.

The output voltage is determined by the rate of field collapse and the windings ratio between primary and secondary. Because the path to ground for the current involves the spark gap, the initial resistance is extremely high. This allows the voltage to build to a very high value until it gets high enough to jump the plug gap. The potential difference must be high enough to first ionize the gas between the electrodes. The ionized gas creates a conductive path for the current to flow. At this point, the arc jumps and current flow is established.

It is important to note that if only 10,000 volts are required to jump a plug gap under a given condition, that will be the maximum delivered. It is also important to note that the spark duration is determined by coil inductance and total resistance of the circuit, plus spark plug gap. Most inductive discharge systems have a spark duration of between 1 and 2 milliseconds.

As cylinder pressure increases, the voltage required to jump the plug gap increases. This is especially true in turbocharged engines under boost. The second problem on high performance engines with high rev limits, is that there is less time to charge the coil with increasing rpm. As such, a high rpm, high output, turbo engine puts greater demands on the ignition system than does a 5000 rpm naturally aspirated engine. Additionally, with a single coil, the more cylinders that you are firing, the less rpm you can run before the spark voltage becomes insufficient to jump the plug gap. A V8 engine would only run to about half the rpm that a 4 would before encountering misfire.

Coil Charge Time and Saturation

The amount of time it takes to charge the coil or bring the current to maximum in the primary windings is called charge time. Input voltage and coil resistance are the main parameters relating to charge time. When the current has reached its maximum value in the primary, it is said to be fully saturated.

If current is applied longer than the time needed to fully saturate the primary, energy is wasted and there is nothing more to be gained. If the current is cut off before saturation is achieved, the maximum spark energy available will be reduced.

Typical coils require charge times of between 2.1 and 6 milliseconds. Obviously, a coil requiring 6 milliseconds to saturate would be unsuitable on a high revving engine as there is not 6 milliseconds available to charge it between discharges at high rpm. For this reason, most performance and racing coils have low primary resistances between .5 and .7 ohms and are fully saturated in less than 3 milliseconds. This permits full coil output at very high rpms.

Most 4 cylinder engines below 200hp/L specific output will run fine below 9000 rpm with a good inductive discharge coil setup.

Capacitive Discharge Ignition

On very high output engines, especially V8 and V12 engines, a single inductive discharge coil is inadequate to supply spark at high rpm and high cylinder pressures. This is where the CD ignition or CDI is used to reduce charge times. The MSD line is very popular worldwide, especially on American V8 engines fitted with distributors.

In normal inductive discharge coils, only 12-14 volts is available from the battery to charge the primary. The CDI charges capacitors to store a high voltage kick to fire to the primary side, putting between 30 and 500 volts onto the primary windings which reduces the charge time substantially. A coil that would take 3 milliseconds to become fully saturated with 12 volts is now fully saturated in less than 1 with a CDI. The same engine now will be able to turn twice the rpm and experience a major increase in cylinder pressure before encountering misfire.

A slight drawback to CDIs are their shorter spark discharge times although it is better to have a shorter spark rather than no spark. One other concern when using a CDI and a distributor especially ones having closely spaced wire terminals is the possibility of crossfiring. This may happen when the coil voltage is so high that the spark will jump to adjacent terminals which can be very destructive. Most high output CDI systems will also run a larger diameter cap to reduce this possibility. Ignition rotor life may also be somewhat reduced.

Some CDIs also include a multispark function where more than 1 spark is generated after the first spark. This improves ignition probability but it is usually discontinued above 3000 rpm because there just isn't enough time available to make this useful. If the first spark didn't ignite the mixture at 8000 rpm, the 2rd spark a few milliseconds later would light off the mixture very late, leading to PCP occuring late with little useful power being delivered. Igniting late is probably better than not at all though.

One company who makes CDIs claims that their system cures all misfires among other dubious benefits. This is a physical impossibility as we have seen above that many factors could contribute to a misfire which are totally outside the realm of the ignition system. An over rich or over lean condition or broken parts cannot be fixed by ANY CDI system.

Many CDIs also claim increased fuel economy which is unlikely as well at the normal air/fuel ratios run in automotive engines. Besides the high rpm coil saturation advantages, perhaps the only other one would be greater resistance to plug fouling. However on modern, well tuned engines in proper repair, plug fouling is really a thing of the past anyway.

Direct Ignition

Commonly known as DIS. Most DIS units are of the inductive discharge type. They use a double ended, isolated coil which fires one cylinder on the compression stroke and one on the exhaust stroke simultaneously. This is referred to as a waste spark strategy.

The advantages of DIS are the elimination of the distributor and the associated rotor to terminal air gap and moving parts, plus the addition of twice the number coils so that one can be charged while the other is discharging. This feature allows DIS to produce a very powerful spark up to around 10,000 rpm.

The disadvantage of the waste spark strategy is that the coils are firing at twice the frequency needed which reduces the charge time window at extreme rpm. DIS systems will usually fire the plugs on any engine up to 10,000 rpm and 300hp/L specific output.

Coil on Plug

The latest, greatest ignition, is the coil on plug setup (COP). This method uses a small inductive discharge coil clipped directly to each spark plug. It eliminates plug wires entirely and does not usually use the waste spark strategy so it has twice the amount of time available to saturate. This basically doubles the RPM capability of the system over other ignition systems. This is the system used in FI and Indy Car engines which generate outputs of over 300hp/L and 16,000 rpm.

Many new top line production engines are starting to use COP.

Ignition Wires

The purpose of the ignition wires is to conduct the maximum coil output energy to the spark plugs with a minimum amount of radiated electromagnetic interference (EMI) and radio frequency interference (RFI). On most street applications using digital computers for engine management control, excessive EMI and even RFI can interfere with ECUs and cause running problems.

There are 3 basic types of conductors used in automotive applications: Carbon string, solid and spiral wound. Most production engines come equipped with carbon string or spiral wound. The solid core types are used exclusively for racing, mainly with carbureted engines because they offer no EMI or RFI suppression. They generally have a low resistance stainless steel conductor. These types are rapidly losing favor, even in racing circles.

The carbon string type is the most common and work just fine in most stock type applications. The conductor is usually a carbon impregnated fiberglass multistrand. Suppression qualities are fine with resistances in the 5K to 10K ohms per foot. They are cheap and reliable for 2 to 5 years usually, then they may start to break down and should be replaced. High voltage racing ignitions will likely hasten their demise.

The spiral wound type is probably the best type for any application. The better brands offer excellent suppression, relatively low resistance and don't really wear out. Construction quality and choice of material vary widely between brands.

NGK makes low priced wire sets which work well in performance and street applications however the terminal ends tend to be a bit fragile.

Magnecor makes excellent quality spiral types with high suppression qualities. These are reasonably priced for the quality you are getting and proven worldwide over many years under extreme conditions.

Some amount of resistance is required along with proper construction to achieve high suppression levels. Resistance is also important to avoid damaging some types of coils and amplifiers due to flyback and coil harmonics. Beware of wires claiming to have very low resistance. These CANNOT have good suppression qualities.

Beware of any wires claiming to increase hp. Ignition wires CANNOT increase hp. As long as the wires that you have are allowing the spark to jump the gap properly, installing a set of $400 wires is strictly a waste of money.

Lately, some truly "magic" wires have come onto the market claiming to not only increase power but also to shorten the spark duration from milliseconds to nanoseconds. As we have seen above, spark duration is determined primarily by coil inductance and coil resistance so these wires CANNOT shorten the spark duration by the amount claimed. The wire resistance has a minimal effect on discharge time because of the high voltage involved. We have also seen above that a very short duration spark is in fact detrimental to ignition because of lower probability.

These same wires claim to increase flame front propagation rates and the ability to ignite over- rich mixtures for more power. We have again seen that once ignited, the mixture undergoes the flagregation process and that the progression rate of the flame front is totally independent of the spark. We have also learned above that most gasolines will not ignite nor burn at air fuel ratios richer than 10 to 1, period, and that maximum power is actually achieved at around 12-13 to 1 AFR so the second claim also has no basis in fact.

These wires use a braided metal shield over the main conductor which is grounded to the chassis. This arrangement offers poor suppression because it does not cover the entire conductor. Any energy leaking out of the main conductor by induction is actually wasted to ground and will not make it to the spark plug. These wires also have very low resistance which as we have seen above, can have a detrimental effect on coils and ignition amplifiers due to severe flyback effects which are normally damped by circuit resistance.

Other claims for these wires include current flows of up to 1000 amps. The current flow in the ignition circuit is determined by the coil construction and drive circuits, not by the ignition wires. Most ignition systems are current limited to between 5 and 15 amps. The most powerful race systems rarely exceed 30 amps. To flow current at 1000 amps, you would require #0 welding cable for the ignition system!

For more info on wires: www.magnecor.com.

Spark Plugs

The last part in the ignition system is the spark plug itself. The average plug consists of steel shell which threads into the cylinder head, a ceramic insulator, an iron or copper core leading to a nickel or platinum center electrode and a ground electrode of similar material. The spark jumps between the center and ground electrode. Certain special application plugs may have multiple ground electrodes.

Different heat ranges are available depending on application. For constant high power applications, a colder than stock plug is usually selected to keep internal temperatures within limits.

Again, many "magic" plugs come onto the market from time to time expounding the virtues of their incredible new design, usually offering more hp of course. Split electrode plugs are a waste of money because the spark will only jump to one of the electrodes at a time in any case.

You will find that most reputable engine builders in the higher forms of racing use pretty standard NGK, Bosch or Champion plugs with pretty standard electrode setups. A properly selected, standard plug will easily last 25,000 miles of hard use in most engines. A platinum tipped plug will easily last twice as long on most engines. There just isn't any rocket science here. Modern spark plugs coupled to modern ignition systems in a modern engine are extremely cheap and reliable. In most cases, on street performance and even race engines, a $2, off the shelf, NGK plug will work just fine.

Wild Claims

If you see an ad for any ignition system component claiming substantial power increases over stock, BEWARE. Most of these claims are total bullshit with no basis in fact. Even if the seller advertises a money back guarantee, you will still be responsible for shipping the product back to them and probably a restocking fee. These companies rely on hype and unsubstantiated claims to sell their products to a predominantly gullible buying public. These people count on the fact that you will probably not time your vehicle's acceleration before and after installation. The seat of the pants "feel" of "increased" acceleration after installing the latest $400 trick gadget is usually enough to sell most people.

If your engine runs clean to redline with the modifications that you have done, it is very questionable that you will make it any faster by modifying the ignition system. If you encounter a high rpm miss at full throttle, there is a good chance that something needs replacing or upgrading.

If you must buy something, stick to reputable manufacturers making reasonable product claims. Steer clear of any company using hype and hard sell tactics and ones claiming vastly increased power or fuel economy. This just does not happen in the real world.

If you see a company making wild claims in their ads, do everyone a favor in the industry as well as the buying public, report them to the Federal Trade Commission.

Update 12/24/02

Just printing a rebuttal to comments made on a page http://yarchive.net/car/engine_control.html. A writer quoted a paragraph on the article above:
"A high compression or turbocharged engine operates with much higher charge densities and consequently faster burn rates. The high octane fuel permits these rapid burn rates because it has far less tendency to autoignite and detonate under these conditions. As a result, high compression and turbo engines cannot realize their full hp potential without high octane fuel."

The response of the expert was " He has this kinda muddled. Fast burn rates, as induced by modern high swirl combustion chamber designs result in LOWER octane requirements and less spark lead."

The author perhaps does not read English very well despite his apparent higher education. Nowhere does my statement address modern combustion chambers or spark advance, it simply states that all things being equal, turbocharged and high compression atmo engines CANNOT reach their full hp potential on low octane fuel. This is simple fact, not conjecture. You can screw with ignition advance all you want, the fact remains that maximum power will be acheived at a certain timing value. If the octane rating is too low to permit this timing value without detonation, the engine MUST make less hp than it could with proper fuel and advance. This whole article and in fact all Tech articles on this site address performance and racing applications mainly. We all know that advances in chamber design has allowed CRs to be increased on the same types of engines from a decade ago in the stock world. This has limited effect on the racing world on 14 to 1 CR engines and turbo charged applications which always use high octane fuels.

The author goes on goes on to make a comment that "water injection has a power penalty associated with its use."

This statement needs qualification. This is true only on atmo engines, CERTAINLY NOT on turbocharged engines. Water injection permits higher BMEPs with the same fuel octane. Very well proven by Harry Ricardo way back in 1933 and documented in his ground breaking text " The High Speed Internal Combustion Engine". Water injection was extensivey used in WWII and post war supercharged and turbocharged aircraft engines for increased takeoff and combat power. It was later applied on F1 and WRC cars.

Many facts about combustion, ignition, emissions and fuels from this era were either lost, forgotten or ignored by the automotive world decades later. The industry spend billions re-discovering what was already well known and documented before in the aero engine industry. There are several excellent texts including "Power Plants for Aircraft" by Joseph Liston 1953, which readers might find interesting.

02/12/03 Update

Be aware that many Japanese spec engines are designed to run on 98-102 octane fuel in their home markets. These engines will not be able to run the same boost levels on North American 92 octane fuel. Expect lots of detonation or spark retard if you attempt this.


I received another E-mail recently from a reader regarding the rebuttal above and ongoing combustion research. Again,I reiterate, this article addresses real world conditions on existing real world, production engines, not experimental designs running in a lab. This reader discusses "negative work" in conventional chambers where the piston is compressing an expanding mixture. A finite amount of time must pass between the point of spark discharge and the point at which the cylinder pressure exceeds the compression line if graphically plotted. This is the ignition delay period and cannot happen instantaneously in ANY chamber. This is the reason for spark advance. In most engines, only the period after about 10 degrees BTDC will have pressure above the compression line. This pressure at TDC is only a small fraction of what PCP rises to at 20-30 degrees ATDC so while slightly counter productive in one respect is also a necessity in the process. Obviously, if we could speed the combustion process so that we could delay the spark until 10 BTDC and achieve nothing above the compression line until after TDC and still achieve PCP at 20-30 ATDC, we could gain HP and efficiency. This condition does NOT exist in current production, mass produced engines. Research is being done to bring this to reality sometime in the future. Detonation IS a REAL concern in all turbocharged production type engines running relatively low octane fuels.