Daox:
As I stated I've read that BMW recommends this, but so far I have yet to find it as a direct source. I have read it from a gentleman who posts on Dodge Truck forums and lists who has spent much time experimenting with fuel efficiency on his truck and researching efficiency and aerodynamics. The feet/minute number is a rule of thumb as far as I know so it may not be optimized for any specific engine design, but all of us here probably lack the means to determine the optimum number.
I find that the 1200 fpm target correlates well with my personal experience. My engine, with a 3.40" stroke, hits 1200 fpm at 2117 RPM. I have gotten my best fuel efficiency lately shifting at 2100. I get my best steady-cruise efficiency at 2100 in 5th (75 mph) with another "island" of efficiency ~1800-1900 in 5th (65 mph) both achieving identical MPGs. If I wound out to 1500 fpm that would put me at 2647 RPM and I know it's not worth my while to drive that fast on the interstates (would have to put me near 95 mph), when accelerating from a stop up a hill I will let it rev to 2500 before shifting and find that not only does it work better in the elevated torque band, it keeps me out of open-loop enrichment under the load of uphill acceleration.
There is a direct correlation between an engine's stroke and the speed at which that engine will naturally produce a torque peak. As noted above the engine's stroke directly correlates an engine's speed to a mean piston velocity. As engines produce more low-speed torque they need to be run slower to maintain efficiency, and as engines produce more high-speed torque then can be run faster and still be efficient. All you folks out there with Honda b16s with a 3.05" stroke should let 'er rev up to 2360 instead of dogging it down below 2000 rpms (Wiki says the D16A3 in a Civic HF has a 3.405" stroke, putting you guys at 2100 rpms like me).
__________________
|