Throughout the course of selecting topics for this column, we've included concern about sharing not only moderately theoretical information but the practical side of how various concepts and techniques can be used at the weekly racer level. Especially when it's possible to share hands-on information that advances an engine builder's or racer's routine efforts to become more competitive, we've tried to zero in on useful topics. This month's topic fits into that mold. Hopefully, you'll derive some thoughts about some things to try, based on your own experiences.
Among the various aspects of the combustion process, let's spend a little time on two of them; air/fuel charge preparation and cycle-to-cycle variations not only among an engine's cylinders but within any cylinder by itself. In its purest sense, air/fuel mixtures are prepared by either a carburetor or fuel injector. Particularly in a racing engine, substance in the inlet manifold will exist as a combination of liquid droplets, vapor, some form of liquid film or essentially air, depending on the type of inductions system in use. However, at least in the combustion space, only in the vapor form can we expect fuel to be converted into useful work at its greatest and most controllable rate. Poor or interrupted atomization of liquid fuel can lead to lost power, wasted fuel and the potential for excessive engine parts wear.
If a carburetor is the method of air/fuel charge preparation, you're already familiar with many of the deficiencies these parts contain by virtue of the concept. And regardless of the number of throttle bores used, these deficiencies exist and are the object of numerous techniques intended to overcome the inherent problems. We'll talk a bit more about that later. If fuel injection is the method of air/fuel charge preparation, the two basic system designs are single-point (from some central location) or multi-point (one injector/ cylinder). Truth be known, all three of these methods are compromises to introducing fuel immediately into the combustion space, as is currently the basis for direct injection, gasoline-fueled, on-road vehicles (not unlike diesels).
When comparing the two types of fuel injection (single-point and multi-point), there is an interesting factor that comes into play. Single-point fuel injection systems (shown, left) tend to be somewhat akin to a carburetor, in terms of having atomized fuel routed throughout the intake manifold (much like with a carburetor). While there is some advantage to having increased atomization efficiency as produced by a fuel injector (as opposed to far less such efficiency by use of a carburetor), atomized fuel is still subjected to the pressure excursions (unsteady flow conditions), flow directional changes and related perturbations that tend to allow atomized fuel to contain larger fuel droplets and experience mechanical air/fuel separation. Mind you, not to the extent of a carburetor but more than when using a multi-port system.
In the case of a multi-port system, fuel is typically introduced much nearer the cylinders (often at the beginning of inlet ports in the head), thus reducing the influence of factors just described for single-point systems. But here's an interesting point. If air/fuel charge quality (at the time of combustion) is poorer in a single-point system than those delivered by a multi-point system (and they typically are), can power levels be correspondingly different? It's clearly possible. Although a single-point system may be less expensive than a multi-point system, tests have shown there can be as much as a 10 percent increase when using a multi-point system, all else being equal. As you would expect, such differences are enhanced when a multi-port system is sequential (follows the engine's firing order).
Plus, multi-point system intake manifolds can be designed to optimize airflow and possible "tuning" features, largely because these are (by comparison) "dry flow" systems not burdened by the need to move air/fuel charges through the manifold prior to delivery at the intake ports. This is not to suggest that "wet flow" systems (carburetor or single-point FI) can't benefit from conventional intake manifold tuning techniques. On the contrary. However, removing fuel from the air in a "dry flow" environment eliminates one dynamic in the task of optimizing an intake flow environment in which pressure excursions and variations in kinetic energy within the system are in a rapid and continuous state of change.
If we have established the possibility that by moving from carburetors to multi-point fuel injection there can be inherently better control of air/fuel charge quality, it's still reasonable there can be variations among air/fuel ratios (at the time of combustion) among cylinders. For a number of years, by the use of in-cylinder pressure measuring devices (using virtually real time data acquisition techniques), we know variations in air/fuel ratio affects both burn rate and peak cylinder pressure (including at what crankshaft angle peak pressure occurs). Of course, combinations of these two factors can affect net power output, positively and negatively. Thus, it is desirable to make every effort to enhance air/fuel charge equality (cylinder to cylinder) and focus on improved fuel atomization, particularly when using a carburetor or single-point fuel injection system. All this comes down to the matter of trying to minimize variations in combustion cycle-to-cycle pressures, especially within individual cylinders. Failure to do so, or at least attempting to address the issue, invariably penalizes engine power.
And regardless of engine displacement, mechanical compression ratio, style of headers and combinations of other power-enhancing parts or systems, it's critically important to stay focused on the combustion space in an effort to create the highest levels of air/fuel charge quality and equality of charge delivery among all of an engine's cylinders.
Some hands-on thoughts to share
OK, at least in theory, all these perspectives may make table-top sense, but we need to drill down into the everyday side of transferring concepts to practice. So where do we start? The following little exercise addresses several of the concerns described in the previous paragraphs. Give what follows some thought. It's not Computational Fluid Dynamics (CFD), but worth consideration. I've not forgotten what it's like to race on a budget.
It's foolhardy to believe cylinder-to-cylinder airflow balancing can be accomplished on a flow bench, regardless of how air/fuel charges are being prepared (by carburetor or some method of FI). However, it's possible to eliminate several of the more glaring problems by beginning with this approach.
Next, and I've found this to be of some value, mount the manifold to be used on a flow bench, tape off all runners except the one to be measured, set intake valve lift to roughly 65 percent of maximum value, start the bench and spray two quick bursts of machinist's blue dye into the plenum (runner entry). Shut down the bench, remove the manifold and investigate where the dye was concentrated along the intake path. These will be locations where the flow surfaces need to be seriously roughened in texture. The objective is to create eddies in these areas that excite the airflow boundary layer and help promote separated fuel back into suspension. Repeat the process for all other runners in the manifold, and don't expect all locations to be in the same place in each runner. The method isn't fool-proof, but it's remarkably effective.
Once you've dyno-tested the "modified" manifold (optimizing fuel and spark settings), note any downward trends in b.s.f.c. values. If you do, chances are you'll also see some slight increases in power. And even if you don't, b.s.f.c. reduction normally translates into sharper throttle response and improved transient torque output…like off the corners.
If air/fuel charge quality is poorer in a single-point system than those delivered by a multi-point system, can power levels be correspondingly different?
Poor or interrupted atomization of liquid fuel can lead to lost power...