[coverturtle]

I would say some testing needs to be done. It's too easy to generalize and what may be the best way to drive one car may not be true for another.

In my reading about car engine design, I was surprised to find that the design point for car engines is 85% throttle. This means that car engines are, or were, most efficient at 85% throttle. That's probably volumetric efficiency. With the advent of computer control, this may no longer be relevant. From a theoretical viewpoint, the factors we want to look at are instantaneous fuel consumption, speed and aerodynamics. It's relatively easy to hook up some sensors and measure or calculate fuel consumption, speed, distance covered and acceleration over a test run but not so easy to figure in the negative force from air resistance. Finding the most fuel efficient way to drive a car may require a number of test runs, but if the test runs do not give consistent results, it may be problematic.

Recently I drove my car over 200 miles at freeway speeds and found that the mpg of my next tank of gas around town was much better than before the run to the big city and back. I attributed this to the distance run "cleaning out the engine" or burning up the combustion byproducts such as carbon buildup, temporarily making the engine more efficient until the small errands around town builds up the deposits again. It reminded me of earlier days when I lived with racers who would change the jets in the carburetors to give the best air-fuel mixture. The criterion at the time was the color of the deposits in the exhaust pipe. Black was too rich, nearly white was dangerously lean and gray was ideal. How gray? That was a matter of experience and the color had to examined at the end of a good long run. For some that was just 10 miles on the freeway; for others it was at least 30 miles.