> Not directly applicable here except in the sense that we should make no

untested assumptions, but here goes:

Kent Paser of 'Speed with Economy' fame played with 'reversion cones' on his Lyc. They are cones smaller in diameter than the exhaust ports & mounted directly in the ports. Obviously this should interfere with exhaust flow but he found that they can improve breathing in a Lyc.

The NA splitters look suspiciously like flattened cones....

To invert Bob's tag line: when things change, things change.

Charlie
(just stirring the pot a little)

Ok, "Pot Stirrer"{:>).  The reversion cones in exhaust are designed to do just that as you are aware.  Prevent the reflected pressure wave in the exhaust from pressurizing the cylinder while it is open for the intake function and hindering intake of fresh fuel/air. However the only  Mazda engineering data that I could find indicated the primary purpose of the "splitters" are for noise suppression. Here are two charts on the Mazda Exhaust splitters.  The technique appears to be more effective at the higher rpms and as Tracy and I speculated, probably has little effect on power until the higher rpms.  From the chart, it looks like the technique is really effective after 4000 rpm.  So if it is at the same time affecting power, it may be happening at lower rpms than I originally would have thought.  Here are the two charts.

7-207.gif shows the splitters  and 7-208 shows the 8 db noise reduction obtained by use of the splitters.  I believe the tuboblock has no splitters because it has the turbo to deaden the exhaust noise - if the splitter were for reversion prevention then I would think the turbo blocks would have it as well.

Not conclusive proof, but the only data I could find on the function of the splitters.

Ed A