"How the hell do you think that's going to make a difference," is my recollection of Smokey's reaction the first time he and I discussed certain ways of improving fuel atomization and the combustion process. Challenging as this might have seemed at the time, we eventually came to some conclusions you may find worth considering. Some of these notions have been touched upon in prior Enginology columns. But even though the majority of circle track engines are carbureted (EFI is on the way, however), most of the following paragraphs will contain information that fits both.
To launch this discussion, let's review a couple prior points. First, there is a clear distinction between airflow quantity and quality. In carbureted engines, fuel typically passes over a longer path to the cylinders than for EFI systems. However, both environments can benefit from providing air that both helps support good fuel atomization and aids fuel suspension (minimizes mechanical separation of air and fuel). It's important that inlet air paths and motion within the combustion space, at the minimum, don't reduce the efficiency for either of these requirements.
Next, recall we previously shared that fuel particle size relates to flame speed; e.g., the smaller the droplet the less time is required for it to be "processed" by the chemical reaction we commonly call combustion. Since larger fuel particles take longer to "burn" (if you will), net flame speed effectively decreases. And, as you know, both these conditions are unavoidably linked to spark timing (the point of spark delivery as a function of crankshaft angle).
Finally, there is the issue of Indicated Mean Effective Pressure (IMEP) which, for purposes of this discussion we'll define as the "net" working cylinder pressure. What do we mean by that? Simply that IMEP is the mathematical difference between positive and negative pressure, acting on a piston during any given operating cycle. More specifically, it's the difference between pressure on the piston prior to TDC of the power stroke (ignition) and then thereafter during the down stroke.
Believe it or not, all of these conditions and their effects can be impacted by not paying attention to how fuel is conditioned prior to and during combustion. Furthermore, since air/fuel mixtures operate in confined spaces and change movement direction with rapid frequency (generally at the rate of milliseconds), taking precautions to do what you can to enhance the reduction of liquid fuel particle size (often mechanically) can be beneficial to increasing power. Now let's talk about some practical ways you may want to consider accomplishing this goal.
There are various schools of thought on how you should treat wet-flow surfaces, both outside and inside the combustion space. Opinions are often based on experience, although sometimes that experience doesn't include any form of scientific analysis or basis in fact, just a dyno sheet. And even though data of this type is arguably a strong point from which to work, getting results without clear knowledge of how they were obtained can leave you in the dark when attempting to make further improvements.
Stated another way, if you don't know why or how you caused a change, what are the chances to repeat the process or make it better? Take time to figure out the "why" behind the "what," if you plan any sort of scientific approach to making changes.
You've likely read accounts of the benefits derived using wet-flow benches. Other proponents suggest proper pressure profiling of air movement and patterns is the preferred method. Swirl and tumble meters can be helpful. Those with access to CFD and related computer software maintain their approach is superior. And then again, you'll find experienced engine builders simply making changes to inlet paths and combustion spaces (piston crowns included) and using brake specific fuel consumption (BSFC) data as the yardstick. Frankly, I've been fortunate to have, at one time or another, used all these methods and found each to provide its own benefits. Sometimes, end-goals will dictate the best method of analysis.
Cut through it all and you'll probably find that the more experience a person has with connecting the dots between the pre-dyno analysis and specific dyno results, the better each method seems to become. As a sage old engine builder I once knew often said, "It ain't what you do, but the way how you do it that counts." Fortunately, all of these techniques can produce worthwhile data that addresses concerns about mixture conditioning.
The region in a flow field nearest surfaces over which a working fluid is passing that is predominantly "stagnant" and dominated by fluid viscosity is the so-clled "boundary layer" and can create significant resistance to net flow. It is also a region where fuel tends to accumnate in air/fuel, wet flow environment. Roughened or dimpled surfaces in the boundary layer region can add energy and help restore separated fuel to the mainstream working fluid, thereby improving the possiblity of iits combusion.
Theoretical pressure distribution in a circular flow passage, steady-state, constant pressure drop across passage. Highest dynamic pressure tends to be at or near the geometric center of passage. The highest point of pressure can vary as a functon of passage and section area shape and section location.
Location and Definition of Boundary Layer in a Flow Passage
Now let's get practical. Inlet air can be likened to a fluid. In fact, Smokey often called it the "working fluid." On many occasions, he talked about studying water movement in creeks just to see "what affected what" during fluid flow.
When air moves across a surface, there's a dynamic interaction between the moving fluid and static surface, amounting to a form of frictional resistance. The smoother the surface, the more random this interaction becomes. Introduce a fluid (in this case a fuel) into the air stream and this randomness can translate into a wide range in particle sizes from the upstream atomization process.
The result? An effective variation in droplet size and, therefore, downstream air/fuel ratios. Power-wise, none of this is good. So while "smooth" flow surfaces may help promote net airflow, a wet-flow environment doesn't always benefit from a fundamental increase in flow. In such situations, creating an increased level of energy at the interface between static surface and wet-flow (at the boundary layer) can be helpful. Best case, you'd like to create a measure of control in the way the boundary layer aids either fuel atomization or fuel suspension (or both). That's where the so-called "rough surface" notion comes into play.
Essentially, without delving too far into an aerodynamic analysis of the interaction between an object moving through (or being passed by) a working fluid, here're a couple of practical thoughts.
A "roughened" surface (ranging all the way from a simply textured to fully dimpled pattern) will create a network of eddies, downstream of the point(s) where the working fluid contacts the surfaces. Even though the time periods of flow during a typical inlet cycle can be very brief and bi-directional, there is sufficient time for the working fluid to become more active along the passage walls.
It's this activity that can aid otherwise separated air and fuel to re-combine and an opportunity to reduce particle size. Regardless of how insignificant this may seem, results (depending upon how badly fuel is not being conditioned prior to these type modifications) can be dramatic. Some seasoned engine builders have been introduced to this concept and have become believers in its merits. Just ask Dennis Wells of Wells Racing Engines
So whether you choose to simply "roughen up" intake passages, selected spots in the combustion space (including areas of piston crowns) or take the time to identify where these locations are the most sensitive, chances are better than good you'll see gains at the flywheel. Smokey even took the idea to the point of developing a combustion chamber and piston crown that created rotational motion of the air/fuel charge, during the compression stroke, but that's an entirely different story about mixture conditioning.