Before we delve into the finer points of EFI table adjustments, let’s take a step or two back and understand exactly what is happening under the hood. It has been said before and it is reinforced here again: An engine is little more than an air pump. It just so happens to be that with the right mix of ingredients, we get to harvest a little hidden energy on the way through. Both the air we breathe and the gasoline in the tank are made up of complex combinations of chemicals with all kinds of hidden potential. However, this reaction can only take place between certain amounts of each chemical. Just like a baker knows that too much flour turns cookies into biscuits, too much fuel gives us a less-than-ideal reaction.

 

 

Ideally, to have no extra molecules of either oxygen or fuel left over at the end of the reaction, we must start with the right ratio of components. Chemists call this correct ratio a stoichiometric mix. For gasoline and air, it is 14.68 pounds of air for every pound of gasoline. Notice that we say pounds of air and not cubic feet. On a molecular level, each string of octane and each oxygen molecule have a specific mass. To get the right ratio of strings of octane to oxygen molecules, we must calculate based on mass. Changes in barometric pressure, manifold pressure, and temperature have a significant impact on the density of air and fuel, so we must remember that one cubic foot of air does not always contain the same number of oxygen molecules. Once again, this is where electronic fuel injection shines with its ability to compensate for such changes almost instantaneously.

 

 From the periodic table of elements, we can see the molar weight of oxygen, hydrogen, and carbon. When these combine to form isooctane or O2 molecules, the mass of each can be calculated from the molar weight of its components. (Nate Tovey)

 

The air you are breathing right now is not pure oxygen. The primary component is actually nitrogen, with only about 23% oxygen. This is why the engine must consume more total air mass to obtain the oxygen required to react with the gasoline. (Nate Tovey)

 

 

4 Cycles of an Engine

“Suck, Crush, Bang, Blow,” or at least that’s how I remembered it in college. Maybe “Intake, Compression, Power, Exhaust” is less offensive, but I doubt it leaves as much of an impression on the mind. The process begins with a relatively empty space inside the cylinder of the engine and an open intake valve. As the piston moves down, pressure inside the cylinder drops below that of the intake tract and atmosphere. This pressure difference is what pushes the air and fuel into the chamber. Since we know that each ounce of fuel only carries so much energy and must be mixed with an appropriate amount of air to burn, the more total mix that can find its way in to the cylinder each time the valve opens, the more potential we have to make power.

increase or decrease the amount of charge filling the cylinder each time. The most obvious method of charge fill control is the throttle blade. By closing off a portion of the inlet tract with a blade, the amount of air available for the next intake stroke is reduced. Mixing a smaller amount of intake air with a smaller amount of fuel yields a smaller power potential, but also resists the tendency of the engine to pick up speed. Conversely, with a throttle blade completely open, the amount of air entering the cylinder can be increased by taking advantage of standing waves in tuned length runners, leaving the valve open longer with changed cam timing events, or changing the pressure differential severely with supercharging.

Actual charge fill is an indicator of how much work the engine is doing at the moment. The more charge filling the cylinders, the harder the engine is working. This work is expressed in terms of “load” or “Volumetric Efficiency (VE).” Load and volumetric efficiency are just two methods used to describe the actual mass of airflow through an engine compared to the theoretical mass flow based on its displacement and speed. The theoretical amount of charge fill is the mass of air that would occupy the same volume as the engine displaces. This mass is found by multiplying volume and normal atmospheric density.

 

 

The theoretical filling is calculated at standard temperature and pressure (STP or STD) where density (rair) is equal to 0.00004671 lbm/in3. To find the flow rate, we normalize for the number of complete displacements over time. A four-stroke engine has two revolutions per cycle of displacement, so the displacement rate is one-half of actual engine speed. The theoretical airflow rate of an engine is:

 

 

The theoretical 100% filling of a four-stroke engine with a 300 cubic inch displacement operating at 1000 rpm is calculated by:

 

(300 in³) x (0.00004671 lbm/in³) x (1000 rpm)

----------------------------------------------------------------- = 7.01 lb/min air mass flow through engine

                                 2

 

 

This engine would theoretically move 7.01 pounds per minute of air through itself at standard ambient conditions and 1,000 rpm. By closing the throttle blade to reduce airflow and horsepower, the engine is only allowed to move 1.40 pounds per minute. The ratio of actual instantaneous mass flow to theoretical pumping gives us volumetric efficiency or load.

 

 

This means that the reduced throttle position yields a volumetric efficiency of 20%. Typical engine loads can vary from 10 to 18% at idle. Most steady cruising at street and highway speeds happens at approximately 20 to 30% load. Light acceleration is usually between 30% and 60% engine load. Wide-open throttle operation for naturally aspirated engines results in a load anywhere from 60% to 105%, depending on how efficient the intake, camshaft, and cylinder designs are. Supercharged engines routinely see loads in excess of 100% once boost is present in the manifold. For reference, most 5.0L Ford Mustang engines show approximately 80% load at peak power. With a Vortech centrifugal supercharger making 8 psi of manifold boost, the same engine can show approximately 140% load at peak power.

 

 Datalogging of the engine sensor data during a wide open throttle run reveals the actual air consumption. This air consumption is used by the PCM to calculate engine load. Notice how this naturally aspirated engine develops approximately 86% engine load at WOT.

 

An early form of fuel injection is central port injection. Seen here, the injectors are mounted in the middle of the intake manifold on what essentially looks and acts like an electronic carburetor. (Nate Tovey)

 

EFI systems excel by being able to accurately calculate engine load at any time. Load has a large impact on what the engine wants for operating parameters to perform optimally. As load increases, spark advance typically has to be reduced to prevent knock. At high loads, fuel enrichment may also be necessary to control exhaust gas and component temperatures.

Any way you slice it, just pumping air from the outside world into the cylinder takes some amount of energy. The engine needs to come up with some power to drive this process somewhere. Once a mix of air and fuel has found its way into the cylinder, we need to find a way to do something with it. The piston moving up in the bore with both valves closed does two things very effectively. First, it compresses the mixture, giving it a new set of properties. A denser air/fuel mix tends to put all of the molecules in much closer proximity, making for a faster reaction once things start to happen. Also, Boyle’s Gas Law tells us that as the volume decreases, pressure and temperature increase. PV=nRT, remember? A hotter, denser mix makes for a faster reaction. Just like the intake stroke, compressing this mixture takes energy to drive the piston up against a mixture that generally doesn’t want to get any smaller on its own. This energy needs to come from somewhere too.

So where does the power come from? Hydrocarbons in the fuel have a tendency to give off lots of heat and pressure when mixed with the right amount of oxygen from the air in the presence of heat. Engineers and chemists call this an exothermal chemical reaction resulting from the mixture of octane and oxygen with a heat catalyst (pyrolysis); we call it the combustion. This combustion releases heat that dramatically increased the pressure in the cylinder. The real trick for the calibrator is to create conditions in which as much of the air/fuel mix that has already been introduced into the cylinder reacts during the time when the piston is traveling downward. Taking advantage of 100% of the available chemical energy in the charge mix during this time means less wasted energy to be dumped out the exhaust or simply not react at all, yielding better emissions and power output. Once we have had this reaction that releases heat and pressure, the result is a piston that is pushed down the bore giving rotational energy to the crankshaft. If the energy pushing down on the piston in the power stroke equals the energy needed to pump the air through the engine, compress the mix, overcome friction, and rotate the accessories on the engine, speed stays constant. If the energy resulting from the power stroke exceeds the other losses, the engine accelerates. Likewise, less energy pushing on the piston than current losses yields a deceleration.

At the end of the power stroke, we are left with a cylinder full of mostly spent gases. As the exhaust valve opens, the piston drives upward providing enough pressure differential to push things out. Obviously, the greater this pressure differential is between the cylinder and exhaust system, the more flow. Much like the intake manifold, proper tuning here to take advantage of standing waves helps scavenge the cylinder more efficiently, which leaves room for more air and fuel on the following intake stroke. As a side benefit of the relatively high pressures driving exhaust gases out of the cylinder, it often becomes rather easy to increase total system flow by reducing restriction in the exhaust system as a whole. The major tradeoff here is noise. This is also where a turbocharger shines with its ability to harness the high pressure, high temperature exhaust energy for more help on the intake side. We will cover more about this later.

 

Air and Fuel

Let’s take a look at what we’re really moving through the engine for a moment. “Air” is actually a mix of primarily nitrogen and oxygen. The ultimate goal of the EFI system is to accurately determine the correct ratio between oxygen being ingested by the engine and fuel being delivered through the injectors. In the science of calibration, we must make the assumption that our external environment is a relatively constant mix of approximately 23% oxygen by mass (20% by volume). Even as ambient conditions change the density of the air, all we can really do is model total air mass entering the cylinders.

Oxygen is the real player in the combustion event, since it is the molecule that eventually mixes with the hydrocarbon to yield carbon dioxide and water. Yes, water. Keep in mind that under normal operating conditions, exhaust gas temperatures are well in excess of water’s boiling point of 212 degrees F, yet water is still present. It takes relatively cold ambient conditions to actually see steam in the exhaust. On cold starts, the exhaust system itself on most vehicles is cool enough to allow some steam to condense to a liquid. An interesting side note is that repeated short trips where the exhaust is cooled below this condensation point yield a fair amount of liquid water, often sitting inside the muffler. This can sometimes lead to corrosion of unprotected components. 

The primary component of our atmosphere, nitrogen, is mostly just along for the ride during the whole engine cycle as far as power is concerned. However, it is important to note that emissions are directly impacted if combustion temperatures get high enough to break down the nitrogen atoms, yielding oxides of nitrogen in the exhaust. To combat this, Exhaust Gas Recirculation (EGR) is often employed. Recirculating spent gases back into the intake charges dilutes the energy potential of the incoming charge, reducing its burning temperature potential. Additionally, unburned oxygen and fuel molecules have a second chance at properly mixing before being wasted out the tailpipe. EGR is often controlled by a valve directly between the exhaust system and intake tract. Another source of EGR is camshaft overlap. Whenever both valves are open simultaneously, there is an opportunity for intake and exhaust gases to mix in both the cylinders and intake ports. This effect is typically more pronounced at low speeds where ram tuning has not yet become dominant, but it should be kept in mind for later tuning concerns.

 

 Figure 2-1 Heating values for several common fuels. Notice that the energy contained in these fuels is not always proportional to the amount required for stoichiometry. (Nate Tovey)

 

Figure 2-2 Isooctane is the closest single molecule that best represents the properties of gasoline. (Nate Tovey)

 

Figure 2-3 Lambda influence on component emissions. Running either rich or lean of stoichiometry will drive emissions up. Burn temperature control is the preferred method of controlling NOx emissions. (Nate Tovey)

 

With many different hydrocarbon fuels to choose from, what’s the difference? A major consideration is the amount of chemical energy stored in the fuel molecules. This energy value of the fuel is known as its “heating value.” The higher this number, the more total energy is hidden in the fuel molecules. Good fuels by definition contain plenty of this chemical energy potential and can be consistently mixed and ignited inside of an engine. The heating value is expressed in terms of energy per unit mass (kJ/kg, BTU/lb), making it easy to see how burning more fuel mass releases more energy (see Figure 2-1, properties of fuels). Gasoline is a good automotive fuel choice due to its relatively high heating value, acceptable boiling point, ease of storage, and relative ease of production.

Much like the air in the atmosphere, modern fuels are cocktails of varying ingredients. Gasoline itself is not a pure fuel, but the blend most closely resembles the properties of isooctane. (Figure 2-2) As the name suggests, octane is a chain of eight carbon atoms each carrying a pair of hydrogen atoms and one more hydrogen atom to cap each end of the chain (C8H18). This makes for a relatively long molecule with plenty of exposed single bonds. It is the breaking of these bonds and reforming of new ones during combustion that generates the heat energy driving the power stroke. Isooctane has three of the end clusters of this chain moved toward the center, making it a bit more stable molecule, requiring more energy to break.

Fuels with a higher octane number have a stronger concentration of these more robust molecules versus the more easily broken longer chains. Thus, they require more energy to break these bonds, and have more resistance to knock. An interesting side note is that the energy required to break these bonds comes in the form of heat during combustion. Heat transfer from one molecule to the next requires a small amount of time. Even though this is a very small amount of time, more total transfer means more time for it to happen. Adding more heat to the molecules in order to separate their ions means more time is required. This means that the burn rate is slightly slower for “high octane” fuels. This becomes important later when we choose spark advance timing.

We have already covered the basic recipe for combustion, but much like any family recipe there can be variations. Stoichiometric combustion of gasoline occurs at a ratio of 14.68:1, but combustion can occur anywhere between 7.5:1 and 26:1. This leaves a broad range of possible operating conditions. The catch is that this chemical reaction releases differing amounts of heat depending on the mixture. Generally speaking, rich mixtures (excess fuel, l < 0.9) burn cooler with more hydrocarbon emissions, and slightly lean mixtures (excess air, l » 1.05) burn hotter with oxides of nitrogen forming as a direct result. For emissions, we target a stoichiometric mixture to leave as few byproducts either way. (Figure 2-3)

Engineers often refer to the air/fuel ratio in terms of “Lambda” (l). Lambda is defined as an excess air ratio. Lambda is equal to 1.00 exactly at stoichiometric mixture and increases as the air/fuel ratio gets leaner. Most OEM code is written in terms of lambda for fuel control. This gives the calibrator a quick percentage reference of their deviation from stoichiometric mixture. Modern engines are run at stoichiometry (l = 1, 14.68:1 A/F) under most conditions to balance emissions and fuel economy, so calibrators spend a lot of time targeting this ratio. Lambda is a convenient number to work with since its value represents the correction necessary to the airflow calculation in order to return to stoichiometry.

For best power, it has been found that somewhere between 13.2 and 13.4:1 (l » 0.95) is ideal. (Figure 2-4) The slight excess of fuel in this mix means that enough extra fuel has been injected to ensure that as many of the oxygen molecules as possible react with available fuel to generate power. Since the total power limit of most engines is total mass airflow capacity, this means the engine uses every bit of air it digests in the name of power. Often we find that this ratio is simply not possible to run with the onset of knock in forced induction applications. To offset this, a richer mixture is run for safety, but the phrase is coined: “leaner is meaner.” Most naturally aspirated engines with conservative ignition timing exhibit a bell curve of power centered around this ratio. Smart calibrators stay on the richer side for safety in a normal driver or racecar. In some rare instances such as endurance racing, the leaner side of this curve is used to yield the same power with lower fuel consumption. Great care must be taken when attempting this, as combustion temperatures can skyrocket quickly and little margin is left for knock.

 

 

Figure 2-4 Lambda influence on power output. A slightly rich mixture generates best output. Just how “rich” is best will ultimately depend upon the exact engine combination. (Nate Tovey)

 

A slightly lean mixture with yield peak thermal efficiency of the engine and hottest burn temperatures. The exact location of this peak again depends upon specific engine design. (Nate Tovey)

 

For best economy we swing to the other side of the stoichiometric balance with a target of about l » 1.05 (about 15.5:1 A/F). Since fuel economy is targeted under cruise conditions, plenty of airflow is usually available. In this case we look to get just enough power to maintain a constant vehicle speed. To achieve this, we mix just enough fuel that, if burned completely, makes enough power to maintain current conditions with enough excess oxygen to ensure that no fuel molecules go unused. This is best accomplished at the peak thermal efficiency point of the engine. To make sure all fuel molecules have sufficient air to react with, some amount of excess air is used. Going too lean merely exhibits a lean misfire condition in which the fuel molecules get so spread out that combustion reaction from one molecule to the next becomes sporadic. Again, leaner mixes for extended cruise periods yield higher oxide of nitrogen emissions and exhaust gas temps in addition to better economy. This can be seriously detrimental to catalyst life and emissions in general. Lean cruise conditions also make for a larger transition back to stoichiometric or richer for acceleration.

 

These reasons add up to making a truly lean cruise more trouble than it is worth to most performance calibrations. At any given operation point, the fuel requirements of an engine can be described in terms of power. Engineers often refer to Brake Specific Fuel Consumption (BSFC) as the amount of fuel necessary to make the current power. The units for this are self-explanatory: pounds per horsepower- hour. Knowing this number (or at least having a good estimate of it) helps to determine a vehicle’s fuel system requirements. That is to say that an engine operating at full throttle and a BFSC of 0.5 lb/hp-hr expecting to make 500 hp can have its fuel requirements calculated as follows:

 

(500 hp) x (0.5 lb/hp-hr) = 250 lbs/hr of fuel total

 

If this is an 8-cylinder engine with individual injectors, we can estimate required injector size as:

 

(250 lbs/hr) / 8 = 31.25 lbs/hr for each injector minimum

 

This is the minimum fuel requirement to support the intended power level. However, this leaves no room for error and doesn’t work well in the real world. Fuel injectors operating at 100% capacity usually won’t do so for very long. In order to protect against unforeseen conditions (cooler air, more power, richer required mixtures) and give the injectors a little bit of safety margin, it is typical to leave some degree of “cushion.” 20% is ideal, but it is possible to work with less. So assuming we only want to use 80% of the injectors’ capacity, a new recommended injector size is calculated:

 

(31.25 lbs/hr) / (0.8) = 39.06 lbs/hr (»410.9 cc/min) for each injector safely

 

Knowing a required, safe fuel delivery rate of the injectors also points toward a required fuel pump output:

 

(39.06 lbs/hr) x (8 injectors) = 312.48 lbs/hr total fuel supply

 

With a known density for gasoline of approximately 6 pounds per gallon, pump flow rate is calculated:

 

(312.48 lbs/hr) / (6 lbs/gal) = 52.08 gal/hr (»197.4 L/hr) desired fuel pump flow rate

 

Spark Events

Once a cylinder is filled with a mixture of fuel and air and compressed, it holds a tremendous amount of energy. The release of this energy is triggered by the spark event. The spark is generated by closing a circuit on the low (primary) side of the ignition coil. The flow of electrons through the primary circuit of the coil excites electrons in the secondary side of the coil to an even higher voltage. The high voltage looks for a path to ground through the plug wires, distributor, and spark plugs. With a narrow gap and high voltage present, the electricity arcing across the gap has enough energy to reach well over 5,000 degrees F in this narrow gap. This begins a chain reaction of combustion between the compressed air and fuel in the cylinder. How long this arc is present is referred to as “dwell.” Dwell should be just long enough to ensure proper ignition of the fuel/air mix without excessively draining the coil saturation.

The entire mixture does not instantaneously change state and composition. Because the burning of the air/fuel mixture takes time to create the dramatic pressure rise, it becomes necessary to start the process slightly ahead of time in order to make sure pressures are at their peak at the right time. This “leading” of the ignition event with respect to TDC is what determines the amount of time before peak pressure is reached. (Figure 2-5) Adjusting the ignition lead shifts the arrival of peak combustion pressure relative to piston motion. The object is to get peak combustion pressure just after TDC to maximize usage of the power stroke. If ignition happens too late the piston can literally run away from the expanding mixture. Start ignition too soon and the pressure may peak before the piston reaches TDC. The momentum of the engine pushing against the piston fighting tremendous pressures in the cylinder can lead to piles of expensive broken parts. Adjusting this lead time to deal with changes in operating conditions is the art of calibrating spark advance.

 

 Shown here at 1 atm pressure (14.7 psia), flame speed is proportional to lambda. The trend remains the same at elevated cylinder pressures, although the magnitude of the speed is increased. (Nate Tovey)

 

 Figure 2-5 Cylinder pressure versus rotation (time) for various conditions. The blue line represents the static compression of the engine without any power event. The red and yellow lines show pressure rise associated with igniting an air/fuel mixture prior to TDC. (Nate Tovey)

 

Cylinder pressure effects of various spark advance settings. Notice how cylinder pressure increases with increasing spark advance. (Nate Tovey)

 

 Brake Mean Effective Pressure of various modern engines. Normalizing the engines’ torque output against displacement allows for easy comparison of the efficiency of many different engine designs.

 

Under most conditions, an engine operates with many different spark advance values. As long as knock is not encountered, the range can be fairly wide. Somewhere near the middle of this range is a spark advance value for a specific speedload condition that results in the most engine torque output. This spark advance value is known as “Maximum Brake Torque” (MBT). Actual MBT timing for an engine changes depending on speed and load. At high engine loads, the onset of knock may prevent operation at this spark value, but the object is to try and run as close to this as possible to get the most efficiency out of the engine without causing damage.

The pressure rise resulting from combustion is the source of engine power. To compare the efficiency of one engine to another or even different calibrations on the same engine, we can look at the pressure generated inside the cylinder. Since it is this pressure that pushes on the piston creating torque at the crankshaft, it becomes a good comparison point. Normalizing this pressure with overall displacement allows engineers to work with a number they call “Brake Mean Effective Pressure” (BMEP). Brake mean effective pressure works out such that it is equal to shaft (brake) work divided by engine displacement. BMEP takes into account all pumping, friction, and accessory losses within an engine. The same number without taking into account these losses is “Indicated Mean Effective Pressure” (IMEP). Indicated mean effective pressure is closer to the raw average combustion pressure and still serves as a good indicator of how much total energy is being released during combustion. Normalizing against engine displacement allows comparisons of relative efficiency between engines of varying sizes.

 

Written by Greg Banish and Posted with Permission of CarTechBooks