[chrisl]

Originally Posted by Dake

It's generally accepted that drag increases with the square of the speed. For the purposes of the discussion, the result is the same: lots more gas to go a little faster.

Yes it does - and drag being proportional to the square of the speed means that it is quadratic, not exponential. That's what I was getting at there. Also, with cars, you have to balance a number of additional factors - rolling resistance is approximately constant at all speeds for example. This means that the total resistance acting against a car's motion is nearly constant at low speeds, and starts to increase dramatically at high speeds (which is why crawling along at 15mph isn't necessarily the best for gas mileage). You also have to balance all of this against the engine's efficiency curve - lugging the engine at 30mph might cause you to get worse gas mileage than cruising along at 55mph at 2500rpm for example. Nearly all modern cars do exhibit a sharp falloff in efficiency above 70mph or so, though, simply because they're designed with that kind of cruising speed in mind, and as such, the aerodynamics, rolling resistance, engine efficiency curve, and gearing is designed around cruising efficiently at 50-65mph.

(as I said in the last post, sorry about the pedantry, but I'm an engineer with a masters degree in fluid dynamics, so I tend to be picky about these kinds of subjects)