Steady-state fuel economy better by slowing down to a speed than accelerating to it?

[Slopfrog]

I swear my fuel economy, as read on the UG, is better by ~2 mpg if I accelerate faster than I intend to go and then slow down to my target speed vs. accelerating up to the same target speed.

This doesn't make any sense to me. The engines power output is the same and the drag factors should be the same (at least I think they should be.) I spoke with a friend of mine who tries to hypermiling as well and he has noticed the same thing in his truck but not his car.

What gives?

[nbleak21]

ECU ignition timing and fuel curves are the likely "culprit."

I've found that a short "blip" of the throttle while at cruise speeds will advance my timing from it's typical 29-32 degree advance, up to 39-41 degrees, and will continue to stay there while at a constant speed, allowing less fuel to be required for stoich 02 parameters.

[Slopfrog]

Originally Posted by nbleak21

ECU ignition timing and fuel curves are the likely "culprit."

I've found that a short "blip" of the throttle while at cruise speeds will advance my timing from it's typical 29-32 degree advance, up to 39-41 degrees, and will continue to stay there while at a constant speed, allowing less fuel to be required for stoich 02 parameters.

Ah ha! I've noticed a quick acceleration and slow down will also magically gain a couple MPGs, especially if im "stuck" at less efficiency than i expected to be. Thanks for the info, I will start monitoring that parameter more closely on the UG.

[Cobb]

A few scion owners noticed this too. The reason was that it allowed the vvt to reset.

[radioranger]

I noticed the same thing on my truck, could be all cars have a sweet spot because of timing issues etc, supposedly addressed by sensors , but they are not 100% accurate , especially true on older cars I would think, I think they call it sitting on the backside of the torque peak, so probably depends on gearing etc also often wondered about putting in a knock sensor and a switch to trick one of the sensors to advance timing in certain driving modes, arent most sensors variable resistors, except O2 's of course.

[ps2fixer]

I have noticed similar with my cars. My car runs around 36-39 timing advancement.

Taking off too agressivly results in a lot more fule though, seems to be a fine line.

I normally take off 70-90% load according to my ultra gauge, my target typically is 70% in 1st and 2nd, then 3rd on up is 80%. If I need to get up to speed quicker, i don't let off the gas to extend the time in each gear while keeping the loads the same, upto 90%. Seems i'm only using 1/4 throttle to do this, the rest must be to just dump more fule??