MaximumJavelin

I understand the point of DFCO is to increase engine breaking under certain conditions. In all my readings I'm consistently seeing that "more DFCO spark retard leads to more engine breaking". My question is why does it matter? Once DFCO enables, the injector PW goes to 0 there's no fuel for the spark to ignite, correct? Is it in case there is any residual fuel that's re-entering the cylinder through the exhaust valve opening, that igniting that mixture closer to TDC causes additional "suction" for the pistons to fight against? Even if unburned fuel is reentering the cylinders, I'd think it would be minimal after a few RPM, and certainly during a sustained coast down.

Additionally, when exiting DFCO, you can see how quickly timing is advanced back to the normal table and adjust it with the DFCO spark blending parameters. I assume the point of gradually adding timing back rather than all at once is to eliminate a momentary lean condition in and possible jerky-ness. If you exit DFCO part way through an injector duty cycle for one particular cylinder (which I'd think would be almost guaranteed to be the case), and the injector is commanded to fire but only completes 50% of it's duty cycle because it started late, but spark advance was commanded to be it's normal operating value of say 45* (coming out of 8* or so in DFCO), you could get a nasty knock since the actual AFR in that cylinder is no where near what the PCM thinks. So instead of a 100% blend back into the main spark table, the exit blending table adds timing back at a certain rate depending on TPS position.

My thinking leads me to consider if the main purpose of DFCO spark retard is not to assist engine braking, but to ease the transition out of DFCO in a safe manner for the engine, without creating a momentary super-lean condition that could do some damage.

Am I way off base here? Spark timing theory and application is pretty cool and I'd love to learn more about it's use in this area.

Injuneer

Just thinking out loud....

I honestly don't know for sure, but I would suspect the primary purpose for DFCO is to minimize emissions, secondary purpose to help fuel economy, not to provide engine braking. Why pour raw fuel into the cat(s)?

Maybe the spark timing retard is there to smooth the transition as the fuel cuts back on at low RPM/closed throttle. With stock programming, the transition is fairly seamless... you have to focus to feel the point where the fuel cuts back in.

shoebox

I would say the spark parameters are there to provide driving smoothness.

MaximumJavelin

Serves me right for posting before having my morning coffee. At least I got it right once.

I agree that the main purpose is to reduce emissions/improve economy. But, I'm not really sure why, and this is my main question, how significant spark retard affects drive-ability (including increased smoothness)? I'd guess it only really matters during the exit DFCO transition, since once DFCO is entered an there's no fuel injection, so if spark retard matters at this point, why?

I've done a pretty good job convincing myself that it's primary purpose is to avoid an instantaneous super-lean power stroke for the cylinder that is in the intake stroke during which DFCO exits and gets an incomplete injection pulse.

I find it to be perfect, I can't tell when it engages even when I've set DFCO delay above 0. Maybe I'll set it to something ridiculous like 2 seconds so I can really get a feel for it, but both on and off transitions are transparent to me.

So the idea seems to be that maintaining fully advanced spark would cause a noticeable jerk when exiting DFCO with the torque converter engaged (or manual transmission)? What would cause this? If I go from 0 TPS non-DFCO to >0 TPS, which has maintained spark advance with a minimum of injector pulse, I don't notice any jerkiness when applying throttle. Is the difference between "minimal fuel" and "no fuel" that significant to noticeable drive-ability?