Well, going by this collection of brake-specific fuel consumption curves, peak efficiency is usually around 2k-3k RPM and 60%-80% torque, depending on the engine. My car (a Fiesta) has an engine similar to the top one on the list. At highway speeds in top gear (around ~3000 RPM), I don’t have much spare torque, so I’d guess I’m pretty near the region of maximum efficiency.
All of those BSFC curves are for comparatively small engines however. For a car with a comparatively overpowered engine, I suppose that the regions of maximum efficiency and maximum economy might be further apart.
I think that’s the hang-up here. If your Fiesta is anything like my Neon, at cruising speeds you only have the throttle open a tiny bit. again, you are not going to get anywhere near even 60-80% torque in most highway cruising situations. you just aren’t into the throttle enough to get there.
and how are you determining that you don’t have much “spare torque?”
There is one big thing we can still do to make cars dramatically more efficient and that is to make them impossible to crash. Much of a car’s weight is there to protect the driver in the event of a collision. If self driving cars advance to the state where accidents become statistically insignificant, then we could build cars that are drastically lighter than we have today.
I would like to point out that there are some modern engines that have enough computer power that they run with the throttle plate wide open in the mid range and engine speed with a combination of timing, cam timing, and injector duration.
So on these engines there are no pumping losses in the mid range.
A minor point. You can design a car to have its overall best/peak MPG at a given speed (though due to reality it would probably be somewhere in the 40 to 60 mph range). You can also design a car to have the best MPG possible for a chosen speed. These two things are not quite the same thing.
I should have clarified that I meant a vehicle designed to be run at low speeds, since the OP was asking for a theoretical maximum. Given a cruising speed of 5 MPH, one could design a vehicle which was very fuel-efficient indeed. Given a cruising speed of 1 MPH, one could design another (slightly different) vehicle which was more efficient yet, and so on.
Cruising at highway speeds, stomping on the accelerator doesn’t give much acceleration, and it’s similar in magnitude to the deceleration experienced while coasting at that speed (with clutch disengaged). Thus, the engine is producing roughly half as much power as it possibly can (at 3k rpm).
And, similar to what Rick mentioned, my car has an electronic throttle. The accelerator is interpreted by the computer as a request for torque, and throttle and all the other parameters are adjusted by the computer. Which might resolve our misunderstanding.
It seems I’ll have to back up this impression with math… perhaps later I’ll look up the rolling resistance and coefficient of drag and see if I can calculate the actual power needed for cruising at highway speeds.