Engine-off experiment
 

I've seen the experiment that measured the additional fuel cars use when starting which concluded that 0.2 seconds is all it takes for it to be worth it to turn off your engine. However that experiment didn't take into account the alternator load which I think is the most important factor for this to be worth while. Personally, I have found that I get worse FE when turning off the engine whenever I can compared to using engine-breaking in high gear so the fuel injectors shut out. I am curious about how this time may vary for different cars. My experiment idea is this:

With the car stationary in neutral and close to empty run the motor on idle until the fuel light comes on, then turn off the car. Add a 5L container of petrol to the car (with exact weight measured on a scale for comparison) then run the car till the light turns on again and measure the time it takes to do so. That is the baseline test. The next test uses the same procedure except having the car on for two minutes then off for thirty seconds repeated until the light comes on. The next test is the same but with the off period reduced to 15 seconds off and 2 minutes 15 seconds on. Then the next with 7 seconds on 2 mins 23 secs off. Finally the baseline test is repeated. The time to the fuel light is compared to see which method is the most FE. I might also check the battery voltage at the start and finish of each test to see if there is a drop with all the starting and stopping. I figure for 25L ($30) I will have a pretty good idea of how long you need to turn the car off for. Am I right in thinking that with the car stationary the light will come on at the same fuel level every time? Can you see any other holes in the method?

I wanted to make the test simple and cheap so that different people can do the same test to confirm the results in different cars/ environmental conditions. Would anyone else be willing to do this experiment?

Edit: Oh I forgot, the time between runs would be some constant, say 3 mins so that the starting temperature was the same each time too...

How do you figure? I'm pretty sure the car's alternator was connected during the test. Maybe I'm misunderstanding your point.

But that doesn't make sense. In both cases (fuel cut-off state in top gear vs. engine-off in neutral) the engine is using no fuel. Yet you'll travel a shorter distance when engine braking in top gear than you will in neutral with the engine off. Whichever method takes you farthest in either case will yield maximum fuel economy overall.

doing the baseline again at the end is important because the gas in the tank will expand some as everything underneath gets hotter in a stationary test. i just found today that the injectors on the gas engine do not shut off right away and some times don't shut off at all on coast.

Gauge Accuracy
 

I also have to mention that fuel gauges are very inaccurate. To have the light come on may mean an non-level situation and a variety of other variables. Aside from those, the gauge is more of an approximation.

RH77

It seems I misread/misremembered the experiment, but the higher idling is partly to do with the additional alternator load so i think it's a bit of a stretch to suggest the 0.2 second rule rather than the 7 second one. Its not that I don't believe it, it just that my fuel log seems to be defying this rule, the few times I have gone from shutting off the engine all the time to just engine braking and my FE increased mysteriously. Maybe that when I use EOC I am less cautious with the rest of my driving. I just wanted to do an experiment in my own car to make sure this was true. It also makes it easier for me to explain this to other people. I'm not trying to rock the boat just get more data to confirm the results.

They are both not using fuel but with engine braking the battery and accessories are being charged/powered by the cars decreasing momentum. It also doesn't need energy to start up again and recharge after losing the amps to start it. In my case the engine then needs to re-vacuum the vacuum reserve and re pressurise the power-steering. Most of my driving is on gently hills/ between traffic lights so I'm usually either accelerating up to speed, slowing up to a traffic light or coasting down a hill (requiring braking to not speed). I'm a pretty big fan of engine off at lights I just don't understand how changing nothing else gave me a FE boost.

Didn't someone do a test on this site that concludes that the engine HAS to use fuel in neutral to keep the engine from stalling because the drive line is no longer spinning the engine, so in neutral the engine uses fuel to keep it running? If I'm not mistaken, but I remember someone doing a scangauge test and coasting in neutral uses so much fuel per hour and coasting in gear doesn't. Correct me if I'm wrong though.

That is correct. Neutral is driveline-disconnect, so it requires fuel to keep things moving. ~0.5 GPH for my car.

From my shop manual, "Fuel Cut-off Control: During deceleration with the throttle valve closed, current to the fuel injectors is cut off to improve fuel economy at speeds over following rpm: B18b1 engine: 970 rpm."

Experiments have confirmed.

RH77

In a manual transmission, this seems right. In an automatic transmission, isn't there a small difference in the amount of fuel used after the coast? When restarting the engine after an EOC, doesn't the engine use an extra bit of fuel to start? When resuming normal driving after engine braking, this wouldn't be necessary since the transmission is already in gear and the engine is moving. Granted, this should only have a significant effect for very short EOCs. In fact, if the .2 seconds is a correct number, my whole argument is pretty irrelevant. :rolleyes: Anyone doing .2 second EOCs? :D

I'd love to see numbers on how much fuel the starter uses.

Right now I EOC only when I'm coming up to a red light or other slowdown, and I'm bump starting in 4/5 gear, even if it means idling at the light, simply because I don't want to kill my stater (on my 3rd in my truck, which isn't a daily driver, just don't want to replace another starter for a long while). Also, got to keep the cat warm :D

I'm looking at people's idle readings from their scangauges and even if you're usign 1 GPH in idle (very high), that translates to 0.003 gals/sec. At 0.5 LPH, you're using 0.002 gal/min. Glide 500 mins to save a liter? Doesn't seem worth it.

I've seen lots of empirical data on here suggesting that it works. Maybe the SG misreports idle consumption? Maybe they mean GPM?

NO test needed, if you want the engine to run when it's disconnected from the tranny, you'll need to use fuel ;)