Those familiar with automotive engine design are also familiar with the performance boost which can be achieved with the proper design of the engine manifolding. In other words, the length of the manifold runners, either on the intake side or the exhaust side, affects engine output torque as a function of engine rotating speed. A slight supercharging effect, then, is achieved without the cost of a supercharger.

Since my first exposure to this phenomenon, in the 1950s, I have been intrigued by the “something for nothing” aspect of tuned engine manifolding, wondering whether the phenomenon could be separated from automobile engines and used in other applications.

I quickly realized that much design effort was based on an incorrect application of engineering technology and that more effort should be expended on an understanding of that which is called “water hammer.”

The static and dynamic pressures of a flowing constrained fluid can be controlled in two ways: Either the flow cross-sectional area can be changed (the Bernoulli effect) or the flow direction can be changed (the inertial effect). In the latter case, the situation where the flow velocity is suddenly reduced to zero (i.e., water hammer), the pressure change amounts to the product of fluid density, flow velocity, and the velocity of a sound wave in the fluid. With a flow cross-sectional area change, pressure variation is limited to half the product of the fluid density and the square of the flow velocity.

The water hammer effect, then, is always greater than that available with a flow area change and it is upon this fact that I will develop a system to provide useful work. I will use atmospheric air as the fluid and it will be used in the same manner as that air which passes through a “tuned” intake manifold, so I will not be violating any principle of thermodynamics.

Consider a tube with a constant internal diameter and valves at each end. The tube length will be dependent upon the response time of the valves and the tube diameter will be dependent upon the rate at which energy is to be withdrawn from the system. The tube is charged with air to a pressure slightly higher than ambient. This is a “one time” step and will not be repeated as I describe the system operation. One valve (hereinafter described as the “outlet valve”) is suddenly opened. The other valve (“inlet valve”) remains closed. At this instant, a pressure front begins to move toward the inlet end of the tube at sonic speed. On the outlet side of the front, the pressure is ambient; on the inlet side, the air remains stagnant at the charge pressure.

When the pressure front reaches the closed inlet valve, its velocity direction reverses and the pressure on the inlet side of the pressure front drops by the “water hammer” amount, as indicated earlier. Since the pressure drop is limited to 14.7 psi, the initial charging pressure and resultant flow velocity are not high.

Prompt opening of the inlet valve will result in another pressure front “chasing” the initial front as they move toward the tube exit. If the exit valve is closed after the first front has passed but before the second arrives, the pressure in the tube...when the second front arrives...will rise by, again, the water hammer amount plus a bit more, since the flow velocity will be slightly more than the initial flow velocity. By closing the inlet valve when the pressure front reaches the tube inlet, the air in the tube is stagnant and at the higher pressure.

There now exists the opportunity to release a portion of the air in the tube to drive an air turbine, stopping the release when the tube pressure drops to the initial charging pressure.

Obviously, this is the point which defines the end of the cycle. So, in a small portion of each cycle, air is available at a pressure higher than ambient.

There are two significant differences between the cycle described and that associated with a tuned intake manifold. The first is that the intake manifold runner has a valve at only one end. This means that a portion of the cycle involves a flow reversal.

The second difference is that the intake manifold must include more than one cycle, otherwise, the runner lengths would be too long for containment under the hood. Jaguar conducted dynamometer tests with manifolds operating at only one or two cycles, obtaining engine volumetric efficiencies over one hundred percent.

With the use of a second valve and the limitation to only one cycle, overall efficiency and effect should be more pronounced in the system described. The valves would most likely be solenoid activated and electronically controlled. There would also be one or more pressure transducers.

Though the nature of the cycle might remind the reader of Walt Disney's movie about “Flubber,” the energy change is not all that mysterious. Energy is lost as air flows after the initial charging, but energy is gained as new air flows in. While the energy lost is greater than the energy gained, energy becomes available because the charged air flows out due to the pressure difference (Bernoulli effect) and the new air gains its increased pressure due to the compression as it is suddenly brought to a stop (inertial effect). Again, the pressure difference is small, but the power available can be substantial with a high fluid flow rate. And, at the present time, air is a very inexpensive fluid.