The cross-over pipe is for resonance wave tuning. Since this thread is specifically pertaining to logs and snails, I will limit it to just that. You need to look at the exhaust system as a whole. Each little segment of the log manifold, including the snails, the through-transom hose, and even the ports in the heads have a cross-section area. First as the exhaust leaves the combustion chamber where it enters the port as it passes the valve and then makes the transition into the first leg of the log manifold. Second is an area of gaseous expansion where the exhaust gasses transition from the ports into the log portion of the manifold. Another change in cross-sectional area is most likely present in the snails. Then yet again in the exhaust hoses exiting the transom.
As each of the exhaust valves open and the cylinders blow down, a pressure wave enters the exhaust system and begins a chain of wave events. Each cross-sectional area will resonate in its own frequency.
Looking at the first part of the exhaust system, the exhaust port (starting at the valve seat) to where the gasses first expand into the log portion of the manifold. The pressure wave travels to the end of that section. An equal and opposite pressure wave is reflected back and returns to the exhaust valve. These pressure waves are a transfer of kinetic energy between the gas molecules present in the exhaust system. These waves continue to reflect back and forth until all of the energy is dissipated. Each cross-sectional area will resonate in its own frequency. After about five reflections, the wave has lost so much energy that it no longer significantly impacts the system. This is organ pipe theory and is the concert behind tuned headers and exhaust (induction as well).
The timing of these waves, by manipulation of pipe lengths, influences the point at which the negative wave returns to the previous change in cross-section area. By timing this properly, a lower pressure can be generated in the cylinder which pulls harder (scavenging effect) on the intake, boosting volumetric efficiency. Additionally, each cross-sectional area length can be tuned independently.
In a car, the exhaust system is so long that two separate waves can resonate in the same length of pipe. The collision of waves inside this same length of pipe creates a high-pressure point. A cross-over pipe provides a point of expansion, in that length of pipe, from which both waves can resonate. Dynamics of the two banks of cylinders working together can further reduce the pressure at that point, helping to scavenge.
Years ago, we determined the ideal location for the cross-over pipe by painting the exhaust pipes. The point where the paint burned off was the ideal location where the cross-over pipe was installed. Now there are refined equations and software, such as PipeMax to calculate this.
It is important to note that optimizing only occurs at ONE point in the RPM range.
So, how does this relate to us in the world of jet boats? It has been said that manifolds can represent a horsepower loss of 10% vs. headers. This is because of reduced scavenging and pumping losses.
IMHO, if you are already working at a disadvantage, everything to minimize weaknesses and maximum strengths should be done to optimize performance. However, in a boat, packaging and heat needs to be addressed. This may not even be feasible for some.
An alternative to a cross-over would be to install idler branches (pictured) on each side. These idler branches would be capped and therefore would not have exhaust gas flow. They would however resonate at their own frequency. If tuned properly, it will produce a low pressure in the exhaust system and aid scavenging. A shorter pipe will show a benefit at higher RPM and a longer pipe at lower RPM. Of course heat will still be of concern.
I have been thinking about this for quite some time, but I keep coming back to the same conclusion. Any max-effort program will use headers. But, food for thought.
Cheers,
Joe
P.S. If interested in this topic, I suggest the book Scientific Design of Exhaust & Intake Systems by Philip H. Smith and John C. Morrison
