Filling and Burning
Performance engine building has as metrics a daisy-chain of three efficiencies. These are guidelines that help keep things scientific in the quest for improved performance. The first of these is concerned with cylinder filling. Volumetric efficiency is its name. Air enters the cylinder, but the usual result is less air volume than if you took the cylinder off the engine and set it out on your driveway! The cylinder has been "charged" with less than its capacity, and it must work with this amount, never getting more. However, the engine can be altered to fill its cylinder more. If we manage to get slightly more air “charge” into the cylinder we call the cylinder "supercharged." Everyone knows about superchargers and turbochargers, that’s what they do. But there are also non-mechanical means, and one of these is chemical supercharging. The use of alcohol, nitrous oxide and other oxygen bearing substances force added oxygen into the cylinder. And there are also ways neither mechanical nor chemical. Wave tuning for example uses the lengths and angles of the intake and exhaust tracts to harness the resonances that can persuade the charge to move, though this approach is more subtle and is not blessed with as many concrete understanding of how it all works.

Once the cylinder is charged, that fuel/air glob must be burned to unlock the fuel's BTU energy. Our second metric, combustion efficiency, gauges how much of the charge gets burned. Because not all of it does. The piston is moving all the time combustion takes place; it is still coming up as combustion starts, and is going downward again while combustion is underway. This movement rapidly changes the cylinder's pressure, discouraging steady flame movement across the cylinder. Other things that affect combustion efficiency are combustion chamber shape, intake tract air speed, and spark plug location. Flatter chambers burn more of their contents than do arched ones; high intake charge speed promotes good combustion because the air and fuel stay more thoroughly mixed as they enter the cylinder; and a centrally-located spark plug causes combustion's flame to reach deeper into the remote edges of the cylinder in less time, resulting in more of the charge being consumed.

And the game still isn't over. The charge has entered the cylinder and been combusted. However, what we’re ultimately trying to make is pressure. Thermal efficiency describes how much of combustion’s heat actually ends up pushing the piston downward. As much as 60 percent of this heat is wasted heating up the engine and going out the exhaust pipe. Maximizing thermal efficiency takes the form of well-shaped combustion chambers that finish their job quickly and waste less of combustion's heat—there is less time for that heat to radiate away from the combustion process into the engine castings, and smaller chambers that offer less surface area that will soak up combustion's heat, improving thermal efficiency even more. Coating the piston top, which is actually the combustion chamber's floor, with a ceramic layer, also prevents heat absorption, this time into the piston, with the result greater piston pressure. Even higher cylinder compression can help, because it makes the charge burn more readily and thoroughly.

So this is the efficiency picture. There are other engine efficienciy measurements as as well, such as mechanical efficiency and power stroke efficiency. But you get the idea. Even before the crankshaft turns, the first few stages of the power-producing transaction each waste some of the precious power that will ultimately be outputted. There are a lot of leaks in the engine's processes and each one affects the one following it because the inefficient outcome of each step ends up being the next step's input, and already there is a loss. Plugging all these leaks—as much as possible—is the art of the high performance engine builder, and helps explain why performance engine building is so much work. Not magic, not mystery, but just plain old work.

Combustion’s Secrets
Combustion within an engine is not an explosion, but rather a rolling flame across the piston's top. It continues steadily, outwardly moving like the ripples in a pond, until all the intaked air/fuel mixture is consumed. The heat produced is turned by the enclosed cylinder into pressure which pushes down the piston. Combustion's flame is the laziest part of the engine's power operation—it takes its own sweet time to traverse the cylinder. A valve can be opened many times a second and the piston reaches speeds of hundreds of feet per second. But the flame's speed never changes. It can't be hurried. Everything is subservient to it. Thus combustion must begin even before the piston reaches the top of the cylinder, so that the burn can start and pressure can be near peak by the time the piston is poised for its rapid descent.

However, because of this head start the piston rises against an already-burning mixture. The thing is, there are always pockets of fuel at the chamber's edges that are the last to burn, and while they wait for the flame's arrival they are still being pressurized and heated by the rising piston. These mixture pockets can reach an ignition point on their own. When that happens, they don't burn, they explode. The result, unlike combustion’s smooth push on the piston, is instead a tremendous shock that can actually break the piston. This shock is called detonation. The real difference is time. Normal combustion is spread over many degrees of crankshaft rotation, allowing the engine to channel the force. But detonation hits intantaneously, spending itself in a fraction of the time, and with that comes damage. And lost power. One of the ways you detect detonation is this power loss. Detonaton robs power because it steals energy that should be taking place over a longish, useful time, and instead pinpoints it into a very brief, potentially very destructive moment.

Beware of confusing detonation with preignition. While they can occur in combination, they are not at all the same thing. There are three differences. First, detonation is spontaneous, that is, the fuel's own heat ignites it. Preignition on the other hand is caused by an outside heat source, usually a hot spot in the combustion chamber. Second, detonation occurs after the spark plug's spark while preigniton always before. And third, the results of detonation are a shock-hammering of the piston (resulting in broken ring lands, typically), while preignition more often melts the piston crown. In motoring's earliest years, detonation was the obstacle to engine power. As gasoline began to be refined, chemical lead was chosen to thwart detonation. What lead did was to insulate the air/fuel molecules from each other, preventing them from rubbing on one another, so that their self-ignition point could be delayed until combustion's flame reached them. That such an inexpensive additive could do so much was a boon to engine power potential. But lead actually stymied engine development. It was easier and cheaper to make engines survive through lead than to improve combustion chamber design. Ultimately, the late 1960s' start of the war against lead proved to be the best thing to happen to the internal combustion engine. It forced its improvement.

The flame front’s speed never varies. No matter the engine rpm: a lazy 1,000 revs per minute or a very rushed 10,000, the burn stll takes its time. Thus the ignition timing before top dead center (BTDC) that works at 1,000 rpm is not going to work at 10,000 rpm. Combustion can't be allowed to extend into the next cycle, and the burn can't be sped up. Something has to give. The solution is to start combustion earlier as rpm increase. If at 1,000 rpm the ignition fired the spark plug at 10 degrees BTDC then by 5,000 rpm it might have to start at 30 or 40 degrees instead. This so combustion has all the time it needs to be completed. Starting combustion earlier and earlier as rpm increase is called "ignition advance". The most familiar advance mechanism to vintage Honda riders is the spring-loaded centrifugal advancer whose flyweights rotate the points operating cam.

How early ignition timing has to be depends on how well the air and fuel are mixed, the air's speed as it enters the chamber, whether the air enters in a steady stream or tumbles and whirls, and many other considerations. The most important factor however is the volume and shape of the combustion chamber, In fact, every combustion chamber design “likes” a different ignition timing. Though misunderstood today due to folklore, the ancient hemi chamber needed as much as 50 degrees ignition timing at high rpm. Gradually, combustion chambers were reshaped, being accompanied by more vertical valve angles, until liquid cooling became common. Liquid cooling allowed the most-vertical valves because airflow over the top of the chamber was no longer important, resulting in 30 degrees BTDC timed engines. Production engines with 12:1 and even 13:1 compression are now common.

Engine modifiers have sought for ways to make older combustion chambers as efficient as newer ones. Adding large amounts of squish, the band of tight spacing around the edge of the piston has helped. It wedges any mixture pockets away from the chamber's edges. Swirl is another tactic, which takes advantage of intake air speed and manipulates it through angles so the mixture is agitated when it enters the chamber. One of the more enterprising solutions has been to add a second spark plug to the combustion chamber. By sending across the chamber two flame fronts instead of just one, this essentially halved the chamber's volume. Compression could then be increased, with the result greater power. A two-valve pushrod BMW winning the inaugural AMA Superbike race in 1976, and a two-valve Kawasaki in 1981, proved demonstrably that the dual-plug idea indeed could make older combustion chamber shapes work far beyond their original limitations. Eventually, even the Big Four manufacturers would incorporate dual plugs into production, most notably in their large displacement cruisers.

MEP, RPM and Horsepower
The framework of an engine’s power output can be distilled down to two measurements: the push of combustion, and how many of those pushes happen in a minute (rpm). That push has a fancy name—mean effective pressure. MEP is the single theoretical pressure which, acting on the piston, could do the same work as all the variable pressures inside the cylinder. Thus MEP is a constant that gauges the strain inside that pressurized container. And it is platform-independent; it doesn't matter whether the engine is a four-stroke or a two-stroke, single cylinder or a V-12. It is concerned only with cylinder stress. However, cylinder pressure isn't alone. Before you can have power, you must also have time, and in an engine, time is measured in rpm. There is an interesting relationship between MEP and rpm—they are fundamentally opposites. An engine that makes its power through a lot of rpm produces a relatively low MEP, and the engine that produces power largely through MEP doesn't need much rpm to do it. Contrary to popular wisdom, rpm is not always the major player in the game of increasing engine performance.

A number of problems arise with high rpm, not the least of which is the valve train; it must be rigid enough and accurate enough to avoid "floating" its parts. Valve float is when the valve has so much momentum that its spring doesn't faithfully return it after the cam lobe has passed. It's a dangerous thing because other parts intermittently share the same space, and conflict is therefore inevitable. The valve train is made to cope by using stronger springs, lighter valves, and less flexible valve train components. But the piston and connecting rod are also susceptible to being moved too quickly. An engine's piston speed has traditionally been a measure of the engine’s maximum safe rpm, though it’s been violently stretched in recent years thanks mostly to advances in piston construction. A modern sportbike has a piston speed well above 4,000 feet-per-minute, what thrity years ago was a racing-only level and now is covered by a three-year factory warranty.

Today piston speed is only a general rule of thumb for rpm. One of the current major rpm limitations is ring flutter. The piston ring doesn’t seal all by itself. During combustion, combustion gases get behind the ring and push it downward in its groove and outward against the cylinder wall. Thus a ring with only a few pounds of radial tension can resist over 1000 psi of peak combustion pressure. The ring actually borrows from that pressure. But this works only if the ring stays on the bottom of its groove. As rpm increase, due to inertia the ring eventually opposes combustion pressure and floats upward on the piston's downstroke at its maximum acceleration point near the middle of the piston's stroke. The piston literally out-accelerates the ring, leaving it "floating" in its groove. This cuts off the path of combustion pressure, and the ring collapses inward on itself against combustion's forces, and fails to seal the cylinder. The resulting blowby reduces lubrication and overheats the piston, and even more cooling is lost due to the ring not being against the cylinder wall consistently. There are a number of ways to prevent ring flutter. Obviously, keeping rpm down is one way. Having a very short stroke is another; even at high rpm, the shorter stroke moves its piston more slowly. This has been the tactic used by Honda in modern times. However, an increasingly common and effective solution and one used in combination with other tactics, is the use of thinner piston rings. Obviously, specially-designed pistons are required as well. Due to their reduced weight, thinner rings build momentum more slowly and can withstand high acceleration without floating. All modern high performance engines use surprisingly thin piston rings.

For the mathematically-minded, horsepower can be said to be a combination of three things. The first is force, which is often measured in terms of pressure. The next part is distance. The engine pushes its piston a certain linear distance The third is time. Horsepower is a product of these three things: force, distance, and time. Imagine that a road crew guy is shoveling dirt. Each shovel-full weighs 13¾ pounds. Sure, that's weight, but it is also force. Our man shovels each load of dirt onto a heap about 4 feet away—that’s distance. And it takes our worker one second to do each shovel-full—that's our time element. Combining these three things gives us 55 ft-lbs per second. You may not recognize it, but now we have horsepower. During the world’s industrial revolution, agrarian concepts began giving way to industrial ones. Scientists sought to measure the machines that would replace the horse, so they actually compared the machines to the very familiar animal. One "horsepower” was thus declared equal to 550 pounds moved one foot in one second (550 ft-lbs per second). Our road crew worker's 55 ft-lbs per second sounds like one-tenth as much, and it is. He is producing 1/10 horsepower.