So, ideally, we'd like to have changes in flow direction at least "act" as if there have been no changes. In other words, it would be helpful if flow passage section areas could be configured (for directional changes) that encourage pressure distribution profiles resembling those in a straight path of flow (you may want to read that again, because it's important). It's additionally important to keep in mind that this same concept, in terms of pressure differential manipulation, also works on the exhaust side of the engine. But we'll get to that in a minute.
Exhaust Systems Here, we'll define "exhaust systems" as they pertain to exhaust ports and directional changes in tubular exhaust manifolds (headers). Although the working fluid in an exhaust system differs in both operating temperature and viscosity from an intake system, favorable passage section design is comparable. Not unlike on the intake side where the "short path" notion applies, exhaust passages respond to the same approach, except at high rpm when flow rates are high and the so-called "short side" path plays less of a role (except at low valve lift). Nevertheless, and certainly at low exhaust valve lifts, you'll find that the fundamental shape we'll be describing in a few paragraphs has merit as it does on the intake side.
Flow Passage Area vs. Torque Output Because section area has a direct bearing on port flow velocity and port velocity is tied in with volumetric efficiency (potential torque output), we need to examine how area relates directly to torque. This is important, independent of port shape.
One approach to examining the importance of section area suggests that at an engine's peak torque (volumetric efficiency), there will be a value for mean flow velocity that appears in virtually any normally aspirated internal combustion engine. Typically, the value assigned to this velocity is on the order of 240-260 feet/second. Of the ways that have been used to verify this concept, building intake and exhaust systems sufficiently "tuned" at a wide difference in rpm has shown this value for mean flow velocity to occur at peak volumetric efficiency. Interestingly, the lowest values for brake specific fuel consumption can also be observed at this point, in the majority of cases.
Regardless, as briefly mentioned in a previous CT tech story, you can use port section area to create this flow velocity at rpm where you'd like the intake (or exhaust) system to provide a v.e. boost. Using this same "tool," you can also design/modify these systems at different engine speeds to help tailor torque curves to specific track requirements. And while the following equation (also previously offered) is a greatly simplified version of a more mathematically complex method of analysis, it is useful for designing/modifying components, selecting intake and exhaust components, or evaluating existing parts packages in a surprisingly accurate way. As previously indicated:
torque peak = (cylinder volume x port section area) / 88200
In this case, "torque peak" (or v.e. boost) is the rpm at which you'd like the intake or exhaust to be particularly volumetric efficient; "cylinder volume" is the calculated swept volume (displacement) of one cylinder; "port section" is the actual section area of the passage in question (don't rule out the option of using multiple-sized passages in a given system for a broader range of torque boost); and the "constant" (88200) handles conversions to allow you to use the English measuring system (feet, inches, seconds, and so on). If you're calculating header pipe section area, don't forget to account for tubing-wall thickness.
We've included this again to emphasize its usefulness and, in fact, direct relationship to the port shape concept that follows. It has proven its merits over a long period of time.