Mark, Put me in for one.
John
I am also very interested. Like BowHunter, I'd like to hear more about the FI plans. HondaMan, could you whet our appetites a bit (just to give us something to think about for the next year). I am curious about the throat size and butterfly/slide arrangement, types of nozzles, etc. I assume you'd try to accommodate existing air intake setups. What kind of benefits are reasonable to expect out of an FI setup? Would it's behavior be fixed or adjustable? How would you set up the interface to make adjustments?
Thanks,
John
The Kaw setups of the past make a good basis to start from, since they are about 1000cc. They had some very small Bosch injectors that will be controllable in the ranges of a 750 like these, with approximately the same throat size. The velocities should be pretty close, due to the undersquare CB750, so mixing won't be a terrible issue. I think it WOULD be an issue with, say the CB500/550, because those throats are much smaller and the engines oversquare, resulting in much more experimentation required.
The general throttle-control approach includes either:
1. some of those old-style flat-plate "injector" carbs that were just a flat slide and one gravity-fed jet, omitting the jet, or
2. a 1-to-4 manifold with one throttle plate, but 4 injectors, timed 2x2 on the 180-sides of the crankshaft or 1x every TDC..
Either version would use the Bosch injectors, as they are proven and available. The timing is simply a varying pulse width, based on engine RPM, with a drop in fuel pump pressure at low engine speeds, probably below 1500 or so, to aid linearity of the "feel" of the throttle. Most engines, ours included, won't idle well without richer idle ratios (as compared to midrange mix), and the injectors are not linear at real short pulses, like at idle. So to get that richness in a controllable way, you have to run a longer pulse and less fuel pressure. Probably some sort of pressure bypass at low RPM would do it.
There is also the issue of temperature. This is where Kaw fell down: they had a poor temperature sensor that would fail with age, leaving a bike that would not start cold or was so cold-blooded that owners were forced to add bypass potentiometers (or other tricks) to start them after a few years (because Kaw quit making the sensors, and they got REAL expensive). To avoid this, the oil temp on a CB750 could be monitored with a surface sensor on the oil tank that could be glued on under the side cover. This can be an inexpensive thermistor with a 300 degree range, which leans the mix as the oil heats up.
This system can be analog or microprocessor-based. Analog is simpler to design, but a little harder to implement the low-RPM bypass stuff. Micros can be easily programmed for different profiles, but the circuit complexities of this approach make for much more expense. I'm still thinking about the differences and which ones would make a simpler-to-implement unit.
An analog unit would have 4 adjustments: idle mix, low-midrange mix, upper midrange mix and WOT mix. These would require a tiny screwdriver to adjust. Simple, but a little less tailorable.
A microprocessor unit would have a lookup table of mix values @RPM that could be changed for different profiles. This would require an external computer or handheld terminal to adjust. Very tailorable, but not every rider is as techno as some of the wizards.
So goes the thoughts....
