Now, and because space is limited in this column, I'll both rewind and fast forward some events I'd like to share. During the mid-70s while I was still involved with Edelbrock R&D, one of the concepts embodied in some of the manifolds dealt with specific runner sizing relative to intended torque boosts at equally specific rpm. Further, we were incorporating more than one runner crossection in the same manifold, intended to provide v.e. boosts applicable to how engines were being operated. An example was the small-block Chevy Victor 4X4 where the inboard four runners (Nos. 4, 6, 3, and 5 cylinders) had larger section areas than the other four longer runners. Each set was torque resonant at different rpm; e.g., one set for off-the-corner torque, the other aimed at mid-track power.
On the pages of this magazine, we've shared periodic thoughts about the relationship between an engine's torque and volumetric efficiencies. Briefly, and to refresh your memory, here's a functional snapshot of that relationship.
Under the conditions of normal aspiration, an engine ingests air not exclusive to piston displacement, rpm, potential restrictions in flow paths (intake and exhaust), and pumping efficiency. Within a span of operational rpm, there will be a point in time at which inlet air flow efficiency will be "peaked," before which there was insufficient rpm to optimize the induction process and after which there is insufficient time to maintain that optimization.
Somewhere in that time frame, among the many calls we fielded from both racers and on-road enthusiasts, I received one from a Peter Vorum. It turned out Pete was a graduate student at Ohio State University and completing his mechanical engineering Masters degree on, of all things, tuning techniques for engine induction systems. Based on some similarities in Pete's work and ours, and the fact nothing proprietary appeared involved, we began sharing ideas and experiences about our work. In fact, we even provided him some data that eventually found its way into his thesis.
If we compare the amount of induced air in a given cylinder (at any point in time) with the volume that would occupy the same cylinder if left open to atmospheric pressure and at BDC piston position, this relationship could be described as the engine's volumetric efficiency, at that engine speed. Typically, v.e. is expressed as a percentage. We've also suggested that there's a relationship between v.e. and an engine's ability to produce torque. In fact, were you to graphically plot the two on a rectilinear grid, they'd look remarkably similar, differing essentially from the representation of pumping losses during running of the engine.
Further, as you would expect, certain intake flow rates play into this issue. For purposes of discussion (and at the risk of oversimplification), we could say that means flow velocity in an intake system varies as a function of rpm, for all practical purposes, in a linear fashion. In application, we find that both intake and exhaust systems behave in this fashion, enabling (by proper mathematical modeling methods) the design of these engine systems to produce volumetric (or torque) peaks at desired rpm. In fact, such methods are a tool by which torque characteristics can be influenced for specific application purposes, such as off-the-corner torque and down-track passing power.
Essentially, Pete was taking an academic approach to modeling induction systems and expanding on material originally presented in the Journal of American Physics and published in 1938, in addition to multiple other investigations since those early studies. By combining the effects of quarter-wave organ pipe tuning oscillation techniques with those involving Helmholtz resonators, Pete was able to configure multiple degree of freedom design capability. In simplified terms, this means his model produced single induction systems that enhance torque specifically at different engine speeds. Such systems enable parts designers (fabricators or modifiers) to configure intake (and exhaust) systems that target specific engine rpm. The method is sufficiently flexible to manipulate torque (v.e.) boosts that are intentionally separated or placed, relative to each other, over a range of engine speed. He's still doing that work, today.
From a hands-on perspective, Pete's approach included validation of the fact an intake or exhaust system will enhance torque output (relative to rpm) by encouraging a boost at a mean flow velocity of 240-260 feet/second. In this context, flow passages that incorporate a constant crossection will optimize their dynamics efficiency if significant mass is moving at this rate when such velocity is achieved, even though this notion may appear contrary to other tuning theories. Over time, Pete's ongoing research has added credibility to his approach, including favorable results from his earlier work with a variety of race teams, including Foyt and similar motorsports groups.
Quoting from Vorum, "In continuing work, I added specific time-average flow velocities that produce maximum scavenging and ram supercharge, when paired with Helmholtz-defined torque peaks. In the quarter-wave model, a designer considers the manifold runner length, one equation, one unknown." Pete's model contemplates the effects of multiple equations and unknowns, producing results that are more predictable to ultimate engine performance. Among them are cylinder volume, port length/area, mechanical compression ratio, valve timing, and target engine speeds where he wants a torque boost.
Further, it's possible to match intake and exhaust torque peaks relative to where you'd like them to relate to each other. Stated another way, you can individually tune intake and exhaust systems in an effort to manipulate a given torque curve for specific conditions. In the case of circle-track engines, this technique will allow tailoring an engine to accommodate track-specific gear combinations, for example. Pete says, "Engine performance is a balance of intake and exhaust pumping work, along with combustion heat release. If the pumping work is improved by good manifold design, the output of the engine can depend less on the combustion output." If you'd like to communicate with Pete, his email address is firstname.lastname@example.org.
Aside from the Vorum approach and utilizing concepts I was involved developing when he first contacted me, there's an additional approach you may want to consider. It includes the notion of separately tuning intake and exhaust systems that embodies an additional feature; multiple passage section areas. Here's a quick view of how this works.
If you agree that the 240-260 feet/second mean flow velocity condition occurs at peak volumetric efficiency (essentially peak torque), this approach can be used as a tool. For example, let's consider a single-cylinder engine. (You can also apply the idea to engines of multiple cylinders.) Suppose you size the inlet runner to reach this mean flow velocity at 2,500 rpm, thus producing a torque boost at that speed. Then adjust the section area of the exhaust passage to achieve this flow rate at 3,500 rpm. In a simplified fashion, you'll have broadened the effective torque range of the engine, compared to what you'd have if both peaks occurred at the same rpm.
While he may not recall today, I once shared this concept with Junior Johnson, after sending him an Edelbrock Victor 4X4 with hand-filed core port runners specific to one of his then-current, big-block Chevy NASCAR engines. As I recall, he complimented the manifold with a set of two-sized pipe headers and a camshaft with two sets of intake and exhaust lobes; one pair for the larger (shorter) area intake and exhaust passages and one for the intake/exhaust set of the smaller (longer) passage area. The combination was an engine that pulled hard off the corners and well past the flag stand. When I approached one of my cam grinding friends about cutting a camshaft based on a similar approach (at the time), he said it probably wouldn't work. Somebody forgot to tell Junior about that. It was a Bristol killer.