What car would you pick?

[theholycow]

Confirming when your car actually uses DFCO is very important...if you haven't had a meter on an injector wire then you can't be sure, and if you're trying to leverage DFCO but not actually getting DFCO then you'll use more fuel than neutral coasting (or a higher gear).

[tradosaurus]

DFCO? What the hell?

[bowtieguy]

deceleration fuel cut off...

when releasing the accelerator at rpms above 1k, most modern vehicles use little or no fuel down to 1k rpms. the drive wheels keep the engine going down to 1k, when idle then kicks in to prevent stalling.

it's like free mpgs for a short time. i use it OFTEN as pulse and glide! on longer glide opportunities, it is more beneficial to coast in neutral however. with AT, if you do use neutral glide, it is wise to rev match before shifting back to drive. or you could just use it at eminent stopping situations.

anyway, you DO have DFCO in that vehicle. i never use neutral coasting in my AT vehicles.

[trollbait]

The usefulness of DFCO depends on the make and model. On my HHR, it doesn't kick in until rpms are above 2000. Then, except for newer engines with better valve control, the lack of fuel going to the cylinders means the engine is pumping air, and engine braking increases.

So for the HHR to hit DFCO, I'd have to be cruising at high speed where the braking will cause too much speed loss, or downshift.

On the other hand, the wife's Sable has an absurd coast ability. There isn't a usable difference between leaving it in drive or shifting to neutral. Except when the DFCO does eventually kick in, the slow down slightly increases. You might not notice it.

New Fords use aggressive DFCO as part of their mpg improvements.

I understand the updated scanguage can show DFCO operation.

[theholycow]

The thing about DFCO is it's never simple, and it's often not dependable or consistent. It's certainly not as simple as anytime you get off the gas over 1000rpm.

Some, but certainly not all, of the additional variables that may be considered by the computer as it decides whether or not to DFCO:
- Built-in delay (I observed this on my 2002 GMC)
- Current gear and road speed (maybe)
- Accessory usage - A/C, wipers, lights, etc (commonly reported on Hondas)
- Recent changes in RPM/shifts (I observed this on the 2008 VW I had)
- Once engaged DFCO may stay engaged down to 1000rpm, but it may not engage if you're at 1100 before it starts. It wouldn't make sense to bother engaging for just a moment, so they may include a margin.

When I first learned of DFCO, I was constantly trying to leverage it and I couldn't figure out why my FE went down. It turned out that downshifting my truck for 5-10 seconds of engine braking cost way more fuel because DFCO was delayed by 8 seconds. In my truck I basically have to be looking at 20 seconds of engine braking for DFCO to beat neutral coasting. Don't forget that during those first 8 seconds it runs at higher RPM but doesn't run lean so it's using more fuel than it would at idle.

The 8 second delay is probably not very common.

The VW usually wouldn't DFCO if I had recently shifted; it preferred to DFCO only after steady cruising. Complicating matters, the VW had a "feature" where it would avoid DFCO and even avoid closing the throttle to prevent excessive engine braking for the sake of stability in low-traction conditions, but it is not well documented at all.

Honda Civic drivers who have directly detected DFCO report that the DFCO floor can raise to 1400rpm or higher with A/C and electrical accessories in use.

A meter connected directly to a fuel injector wire is the most dependable way to detect DFCO. OBDII tools like the ScanGauge and UltraGauge do an OK job, with some display lag, if you watch the open/closed loop indicator (other data has worse lag and/or may be inaccurate). Sometimes you can tell by turning off the ignition; if engine braking increases with the ignition off then you were not in DFCO, but that's only useful experimentally, not as an indicator.

Studied, measured, and applied accurately, DFCO is a very useful tool. If you assume or guess then it's a crapshoot and may cost more fuel.