Concerning the step in the port:

In '92 I bought an RS125 NF4 to race the British Championship. The first thing I was told was not to remove the step as you'd lose a hp. No one told me the reason why. I asked but no one would explain, so I left the step in there.

In '95 I raced the MC28. As we know the NSR has that same oval shaped exhaust port and the step where the port meets the header. After lots of grinding, testing, grinding, testing I was getting acceptable power. This was still keeping the step, albeit not as big as it was as stock. Then I filled the (Harc-Pro) manifolds with weld and set about grinding them back to match the oval shape of the port. Threw the bike on the dyno and BINGO......more mojo all through the powerband. Sweet.

When I went to work in Japan one of the first things I noticed was that the HRC "A" kitted 125's and 250 in our shop had matched manifolds. Not welded and ground like mine, but machined for the job by HRC.

Does size matter? yes! But thankfully in this case bigger is not necessarily better.

As the exhaust gas leaves the cylinder it is necessary to keep it's velocity high. This high velocity helps evacuate the cylinder. There is a close link between the size (area) of the exhaust port at the cylinder wall and the exhaust port where it meets the header. If you match the port to the header it basically means that the exhaust gas loses it's momentum too quickly and can effectively block the exhaust. But if you keep the area closely matched to the exhaust port proper then the gas velocity stays high as it charges into the expansion chamber. The step has little effect at this point.

The step helps to reduce the energy of a returning pulse which could contaminate the fresh mixture in the cylinder while the exhaust port is uncovered. It works at all rpm but more effective at low rpm.

Removing the step by using the matched flanges could cause a loss in power at the bottom of the range as both the suction and plugging pulses in the pipe have increased intensity causing over scavenging and also forcing spent exhaust gas back into the cylinder, but at the top of the rev range the lack of turbulence in the header aids efficiency, which means more power.

To this day no one has explained it to me, but that's my take on it.

The manifolds worked for me. They work for Steve and HRC use them. Who am I to argue?

Oh, and I haven't spoken to Wobbly for a week or so.