The instruction manuals for many MBs tell you to coast in neutral in some circumstances. Admittedly, this is only when tow-starting or hill-starting the car. Did you know that was possible with an MB automatic? You have to do 45 - 50 km/h, and then move it into 'D', but if it doesn't start in a few seconds, should move it back.

later,

Russ

modern fuel injection systems deliver idle fuel quantity while coasting. they must maintian combustion to keep the heat up in the catlytic converter for emissions purposes. Try switching the key off while coasting.. the engine will change tones and sound completely different.

A diesel will shut off fuel completely during coasting. If you turn the key off, it doesnt change tones at all, and as you coast to a stop youll hear it start clattering again as it adds fuel near idle speed. This noise is pretty hidden on an MB, but easily heard on pickup and full size truck diesels.

If load is a factor on fuel flow to the engine, then an engine braking event would put a load on the engine and require fuel to govern it (observe the amount of smoke spewing from decelerating trucks, albeit gearing down intentionally); coasting would at some point cause braking and load the engine; the equilibrium conditions where the crankshaft is in tune with the transmission gear and the tire rotation would be rare indeed in such a dynamic and unstable environment as "driving";

therefore, I maintain that "no fuel" scenarios are idealized, with unknown practical value.

I do not know about any of the other MB's, but my 190E 2.3-16 from 1986 has a fuel cut-off when you take your foot off the gas and the throttle closes, at any speed above idle. When the car slows down to the point where the engine idle speed is approached, about 1,000 rpm or so, the fuel is turned back on, in idle mode. You cannot even tell this is going on. The manual shows there are a couple of microswitches to detect throttle position, and there is some other discussion about this feature and how it works. The idea was to improve gas mileage, and I guess it does something for you. I am not sure how this is implemented with an automatic, but I imagine it works the same way. If the engine speed and throttle positions do not require fuel, I see no reason why the fuel cannot be shut off. If you decellerate quicker than you might like, you tip in the throttle a little and the shut off command is negated. As long as the fuel is cut in and out smoothly I see no reason why this can't work with other cars and automatics as well. Good Luck, Jim

consider a thought experiment: coasting downhill with the foot off the accelerator with an automatic tranny. what would happen if you shifted to "3" or "2"? would the sudden jump in rpm's require fuel? does the tranny affect fuel or is it governed exclusively by the engine? does the tranny bear the load of decelration or is it the engine?

Fuel is flowing to the engine while coasting.

I've already switched-off the engine while coasting and you can defenitely feel the difference in what concerns deccelaration.

After the engine is switched-off the car starts to loose speed at a higher rate than with the engine switched-on. It looks like coasting with a car without injection (does any of you remeber?)
Now, what I don't know is the amount of fuel that's going in (more or less than iddling conditions)

This experiment was not perfomed in a Mercedes, though. It was an Honda, with a Bosch injection system, so I think things would be pretty much the same in a MB. (Remeber: this should not be performed unless you kbow EXACTLY what you are doing since brakes will fade away, steering wheel could lock itself, ...:D )

If indeed the mb has a fuel cutoff switch then your result may indicate that

1) it is a useless feature, as it is hardly ever used.

2) the fuel system on a honda differs radically from mb.

the original post to this thread asked how fuel is used under these conditions, therefore meaning a description of its behaviour whether there is technology to cut fuel or not..if it is not used enough to consitute a pattern of behaviour, then listing mb features that are rarely used does not help us explain how fuel is actually used. I beleive your experiment consitutes a situation where "for sure" no fuel is being used and is quite interesting.