Thinking on and looking at Cam timing events and ways to gain more power than I had previously, I think I've arrived at the conclusion that getting much past 260whp on a rocket III, with the air filter in place, on normal pump fuel, on the 2WDW dyno, starts to get into the region of compromising dynamics to the point it no longer makes sense as a street bike. To make a meaningful amount (+10hp or greater) more power requires taking it away from other areas or incurs maintenance and financial cost more than I'm willing to do.
TLDR: I'm finally going to stop chasing more than 260ish whp. I do think a 280-290 is very possible, but would compromise rideability as a street bike.
Looking at the numbers again and refining some data, I think I've finally arrived at some conclusions:
- The easiest "free" power is a swap from mechanical water pump to electric water pump.
- Parasitic drag/friction could be improved through various techniques and parts replacements. WPC coatings, specialized polishing, better bearings, lighter components etc could all free lost power. The costs for this stuff add up incredibly quickly for rather small gains and most of them incur a more frequent maintenance schedule.
- I was already hitting 3rd Harmonic resonance events on the intake side using my curved stacks and PX500 setup, there's nothing to be gained here. Adding a plenum fills in more power below 5000 rpm, but takes some away over 6000 rpm. Changes in runner length simply move the power around, but no specific length will make more overall.
- There's not much to be gained on cam duration, the current duration already starts to cut into the running dynamics below 3000 rpm and much more would make it a much worse street bike imho
- The head is the main flow restriction. The intake valve curtain area is the main reason I seem to have gotten stuck in the mid 250 range for whp. This could be improved in two ways, adjusting the pinch point diameter/seat cut in the bowl and/or using more cam lift. Cam lift is already restricted by valve guide space, leaving, only the bowl and valve seat adjustments to improve flow efficiency. This is not something I'm interested in doing at this time; perhaps in the future I'll send the head to Millennium Tech to get work done in that area. I think Bob did a good job already and improvements that can be made are very specialized. I believe this to be the cause of the sinewave like curves in my torque output from 4500-6500 rpm. No change to exhaust (I've tried 3), no change to intake (I've tried like 5 setups) smooths it out. Cam change didn't smooth it. More and less ignition timing doesn't smooth it. The only remaining possibility I'm aware of becomes valve curtain/bowl/seat angle/tulip interaction.
- Valve shape and back-cut. The exhaust valves aren't bad, the factory did good work here, but there's a bit of room to improve them by back cutting and adjusting the shape of the back of the valve. The intake valves on the other hand, have big fat tulip shapes that certainly impede flow. This is probably the single biggest area left for improvement on a 265 kit head. By replacing the valve guides with 5mm ID and machining the valves down to a 5mm stem instead of 5.4mm more could be gained, but this is not trivial effort stuff.
- The exhaust, does actually have a little room to improve it's just shy of hitting the perfect 3rd harmonic on exhaust tuning, however, packaging and design make this impractical and this is a small improvement. Truly equal length runners that are 3.175mm larger ID with 595mm primaries paired with an 800mm collector (including the reverse megaphone at 13") are the ideal for over 6000 rpm operation, the reality is the 2.5" collector LMS header paired to a 15" reverse megaphone is so close to perfect it's not worth the compromises to adjust and the existing system provides better power from 2000-6000 than the ideal high rpm setup would. Additionally, building a stepped header design that incorporates all of the adjustments would be an absolutely massive increase in production effort and not reasonable for a less than factory supported build.
In the end, I've learned more about NA power than I ever could have expected since I bought the bike in September 2014, but I know there remains a lot I don't know, thankfully we have people like Neville in the community who are gracious enough to give nudges in the correct direction. In the future, I'll probably build another engine and go all in, starting from scratch on the head and valvetrain, but that's an adventure for future me, after I finish school and have more resources than time and can afford to have multiple heads sent to a specialist to be flow benched with various configurations to find ideal numbers for seat angle, bowl size, valve size, runner size etc.
They now have open range on many of the best twisties and dodging black cows was a *****!
