Laying out the relationships between the engine’s exhaust and intake and how you can use them to your advantage

There are specific functional relationships between an engine’s exhaust and intake systems that can be used, once understood, to the benefit of torque output and how in the rpm range these can be achieved. But first, we need to define volumetric efficiency and then discuss certain aspects of how this aspect of an engine’s performance can be manipulated for gains in torque.

Generally stated, volumetric efficiency is a comparison (in units of percentage) of the amount (mass) of air ingested by an engine on each intake stroke compared to the amount (mass) of air that would fill the cylinder by atmospheric pressure if left open to this pressure with the piston at BDC. We know there are numerous conditions (restrictions in the intake path, limitations in time, etc.) that typically produce v.e. values less than 100% in a running engine. Obviously, the approach is to achieve values as close to 100% (or higher in some cases) as possible.

We also know that peak v.e. and peak torque (both relative to rpm) occur at about the same engine speed. As a result, the shape of an engine’s v.e. curve and torque curve is quite similar. Consequently, some engine builders (or parts designers) like to work on shaping torque curves, know the relationship between both conditions. Over the years, we certainly did.

One other item to mention: There is a significant body of evidence (and research results) suggesting that at peak torque (or peak volumetric efficiency) the “mean flow velocity” in an exhaust or intake passage approximates 240 feet/second, possibly 260 feet/second. One value in knowing this is that a parts designer (or modifier) can adjust the dimension that materially affects flow rate and encourage peak torque (or v.e.) to be at the desired engine speed. In fact, this is a powerful tool, to which we’ll continue to refer in these discussions. That dimension is cross-sectional area of the passage. More on this later.

Now, for purposes of simplification, let’s consider looking first at the exhaust side of a single-cylinder engine. At least initially, this eliminates the effects (complications) that are introduced when we add cylinders and they are, in some fashion, connected to and influencing other cylinders in a multi-cylinder engine. If we first consider that when the exhaust event begins (opening of the cylinder’s exhaust valve), cylinder pressure is higher than in the exhaust and intake systems. As the exhaust event begins and continues, cylinder pressure will diminish such that when the intake valve first opens, pressure in the cylinder will be slightly higher than in the intake path.

At this point, some non-combustible exhaust residue can flow into the intake path, thus tending to contaminate the next fresh air/fuel charge. This back-flow or “reversion” should be minimized in order to optimize net power. And while it is beyond the scope of this column to address ways to reduce reversion, they include reducing reverse flow at the exhaust valve and seat, rate of exhaust valve closure (near its seated position), mis-matches between the exhaust port in the cylinder head and exhaust pipe and other means of reducing reverse flow at the valve.

But let’s get back to our exhaust event. We previously mentioned that there is a body of data, developed a number of years ago that pointed to what has been termed the mean flow velocity of exhaust gas (combustion residue) expelled during an engine’s exhaust cycle. Among other variables, the principal ones are piston displacement, rpm and the cross-section area of the primary pipe.

In fact, a mathematical equation has been developed that encompasses these variables such that, by algebraic manipulation, any one of these can be determined as a function of knowing values for the other two. Here’s the equation:

peak torque rpm = (primary pipe cross-section area x 88200) / cylinder volume

This format will allow you to determine peak torque rpm as influenced by a header of known dimension (cross-section area) combined with a known cylinder volume (piston displacement). By using some algebraic transposition, you can determine the primary pipe cross-section area required to create a torque peak as influenced by pipe size according to the following:

primary pipe area = (peak torque x cylinder volume) / 88200.

These two mathematical formats can be used to either decide at what rpm you would like a header to boost torque (in the header selection process) or to determine header primary pipe size to help boost torque at a specific rpm.

It is worth noting that collector volume can also play into the determination of an overall torque curve. Essentially, collector volume affects torque output below that net torque peak rpm. What this means is that as you add or reduce header collector volume, torque is correspondingly added or reduced below net peak torque. Generally, such increases or decreases in collector volume are made by increasing or decreasing the length of an existing collector. And if a situation arises where you need to pretty much kill off torque below the peak rpm, simply shorten collector length short of its complete removal from the system.

Following are some summary points that you may want to consider when dealing with a header system.