First let me say, great article Mike! And interesting results. Good to see someone stepping up and testing to try and answer the hard questions.
Now I have a few comments.
1) Your American Turbine Impeller cut table (p118) agrees with Seloc for A and B cuts but not AA. You had:
AA Front: 9.25"
AA Rear: 7.50"
Seloc has:
AA Front: 9.094"
AA Rear: 7.813"
That's a significant difference, which is it? Rpm vs Hp curves seem to support a larger diameter than Seloc, but Seloc's been wrong in the past, and not sure about these specs just wondering. The A and B cut match with power, but the AA doesn't.
2) It's hard to draw conclusions about impeller matching without an engine power curve. If your engine power curve's similar to that published in the Nov/Dec issue (p106), there's only about 10-12 hp difference between 6200 and 6700 rpm. That engine has a very flat power curve. A flat power curve engine may benefit more from impeller efficiency than matching compared to a peaky engine. Another thing is your 6700 rpm point. 6700 rpm is on the down side of peak power if your engine's the same as the dyno results in the Nov/Dec issue. Not a good match point no matter efficiency or power.
Then there's the rule-of-thumb that 10 hp with a great setup and light-weight jet's good for about 1 mph...OR LESS. That doesn't seem to jive with your measured speed difference of 5 mph between the AA and the B which would imply a 50 hp change. Neglecting impeller efficiency. If impeller or pump efficiency makes up the difference (4 mph) with the larger impeller, then efficiency may be more important than power matching considering your potentially flat power curve.
I'm not saying one way or the other just commenting on your data.
3) This is a comment for those reading the article. The impeller rpm vs size vs speed data can be deceptive. Why? Because even if you neglect pump efficiency, if you run a larger impeller at the same rpm as a smaller impeller you WILL be putting more power to that impeller. More power to the impeller means more thrust. More thrust means more speed.
So what does that mean? More speed at a set rpm with a bigger impeller's a given no matter the payoff in performance with a correctly matched (and possibly) smaller impeller. Or...the fact that a larger impeller gives your jet a higher speed at the same rpm is no guarantee that the larger impeller will give you better acceleration or top speed.
Someone posted, "I have always believed, the AA was the best. more mph and less rpm."
If you think that's the best...then go with it. Maybe a bit simple-minded but no worries. Why not put an AT 9.5 in your pump? Bigger is better right? If on the other hand you want to consider the concept that impeller matching has merit and is not so simple minded...then read on.
4) Finally, I wonder about test data and cut size. From AA to A, a gain of only 150 rpm? That doesn't seem to agree with AT's pump charts for power applied? Not to say AT's pump data is right or your data wrong. Just an observation. Your numbers from A to B match...similar power and expected rpm gain. But numbers from AA to A. Seems something missing...
jer