Please explain DFCO (Deceleration Fuel Cut Off)
 

Hey guys,

I know that it stands for Deceleration Fuel Cut Off and I also know that most modern vehicles including my Scion employs it because I see it on my scangauge. However what I would like is a definition of how it works.

Meaning, I still here the engine running while its doing it, despite the fact that it is not getting any fuel (Supposedly), how does that work? Why would it do it while decelerating in gear and not while neutral coasting?

And most importantly how does it restart the engine while rolling and without any bump start feeling?

Thank you ..

DFCO can only happen while your transmission is in gear. So, the wheels are spinning the engine. To restart, it just injects fuel since its already spinning. If you were in neutral and DFCO happened (which it can't), the engine rpms would drop to 0 and you would have to bump start or key start.

Some cars have what is referred to as DFCO but is actually Deceleration Enleanment. My car with the Buick 3800 is one of those cars but I'm not sure how common that is.

You're not using any fuel while you DFCO, but it saps the vehicle's momentum to keep the engine turning. You're going to have to burn fuel to get that momentum back. DFCO will give you better mileage than using the brake pedal, but less mileage than driving without brakes and engine-off (or maybe even engine-on) coasting to a stop.

Momentum isn't a problem going down a hill... My question is how come the DFCO turns off while still engine braking? I'll be watching the ScanGuage and it will just change from 9999 to ~60mpg after a minute or two. I have to alternate between engine-on coasting and no-fuel engine braking while on long descents. It's like the Ford engineers in Michigan couldn't conceive of the hills we have here in Colorado. Is there a reason for this? Is there a way to change it?

DFCO is employed because when you are engine braking the effective compression is very low, and any fuel injected into the engine has a very good chance of not igniting and being exhausted with very high amounts of unburned hydrocarbons, as well as damage to the converter when the unburned fuel passes through the converter.

With manifold vacuum of higher than 22 inches, the amount of air in the cylinders is only a small percentage of the amount that would create the designed compression ratio. This is a recipe for misfires.

Determining if your car has DFCO is easy, just let the car slow down in gear until you feel the fuel injection engage and the car will either stop slowing down altogether or slow down at a much lower rate. The point is easy to feel.

regards
Mech

Not maybe. :)

Engine-on coasting is what got me above 47mpg (or below 5L/100km) .
I was using DFCO and braking as little as possible before I started coasting, so the combination of both is inferior to engine-on coasting.


My engine uses 0.5L/H while idling.
So it takes 10H to use up 5L.
As long as I do 100km in those 10H of idling, or 10 km/h, I'm still only using 5L/100km and getting 47mpg.
As long as I'm coasting faster than 10 km/h (6 mph), I'm doing (far) better than 47mpg even with the engine on.

euromodder -

Yes, but I think DFCO has a strategic purpose for us. Going down a steep grade while coasting is not a great idea from a brake-overheating POV. However, if I can get engine-braking + DFCO on the downhill, I can avoid using my brakes and risking dangerous speeds.

If I had a fuel-injector cutoff switch, I could use it for manual DFCO.

CarloSW2

round.boater -

For me that usually happens at a specific MPH+RPM. Have you noted the point where this happens to you?

CarloSW2

Basically the car knows it doesn't need to dump fuel into the engine to keep it going so it cuts it off. If you took your car out of gear it would respond by turning the injectors back on. Also, if your RPMs fall below some pre-set number it does the same.

DCFO used with EOC can indeed give you better FE than EOC alone in certain circumstances. I can give you an example: On my drive home from work there are so many chances to EOC for long distances that the headlights can drain the battery so low I may not be able to restart the engine if I mis-time a light and have to stop. In order to prevent this from happening, I frequently would bump-start the engine at times just to get the alternator to charge the battery slightly. I would do this at lights to make sure I wouldn't be stranded. Thus, burning gas that was getting me nowhere.

Now I have an injector cut off switch that allows me to DCFO at any RPM and will not restart the engine when I take it out of gear. I use it when approaching lights that are red or I know will be red before I can get to them. I'm not wasting momentum, since I use it when I know I'm going to be stopping anyway. DCFO also keeps all the underhood pumps working, including A/C if it's on.

As to why your car drops out of DFCO after a while, there are almost always RPM limits for DFCO. After all, if they left the injectors off all the way down to 0 RPM, you'd have to use the key to start again and that wouldn't be so great.

It is also possible that Ford engineers decided that running in DFCO for more than X amount of time ran some risk they didn't want to run. I'm not sure what sort of risk that could be, though. You should still be pulling a vacuum so the brakes should keep working; with no fuel and no spark you shouldn't be putting a thermal load on the engine (just compression heating which isn't nearly as much as combustion heating); the field coil on the alternator should still be energized so the charging system should still be charging... No, no idea why they'd do that.

-soD

some_other_dave -

My car usually exits DFCO around 2500 RPM in 3rd or 4th gear. I wish it was around 1000 RPM, but maybe that's too close for comfort because idle RPM is about 850.

If the Ford has unheated oxygen sensors, then a long period of time would cool off the sensor and it wouldn't read correctly. A cold oxygen sensor would encourage the ECU/PCM to switch from closed-loop to open-loop mode, which is not good for MPG.

CarloSW2

Going down a rather steep incline, you obviously have no choice but to remain in gear and use engine braking.

If gravity is propelling your car, DFCO is better than coasting with the engine on as there is no fuel use at all using DFCO.

But on the near-flat terrain where I live, engine-on coasting is definitely better for FE than DFCO and trying to brake as little as possible.

Exactly as Euromodder said. DFCO is best used for when you have no choice but to bleed off momentum. At least you can keep the engine from using fuel while doing so, plus it saves wear on your brakes.

The beauty of many of the roads I drive, I suppose, is lots of time to watch the guages.
My DFCO cuts off at ~2000 rpm regardless of time.
Coming down a hill, the water temp will drop to ~192 with engine-on coasting, temps hold ~198 with RPM's in the mid 2000's and no fuel, temps climb to ~206 with RPM's in the mid 3000's during prolonged no fuel engine braking. An 8 mile down grade of consistent 6 or 7% (I-70 tunnel to Silverthorne) will have me cycle between DFCO and engine-on coasting several times. Pulling a trailer I can't really coast out of gear though or I'll melt my brakes.
I'll bet the exhaust gas temps are the reason for cutting off the cut off. Too bad, most of those descents transition into climbs, and temps will jump right back up with hardly any wait at all.

Understanding its purpose and use how come DFCO reads 9999 on the scangauge while EOC reads somewhere in the ball park of lets say 2800 mpg's, why not 9999 as well?

In my daily commute, I've started adding a DFCO phase to the endo of my EO(n)C sections..

this presents the ideal opportunity to slow the car for the ROundabout/junction

Using no fuel and being in the right gear for picking away again!

The OBC shows its a positive step..

fingers crossed it works on this tank!

If you can't start coasting from further out, DFCO is the way to go for the last part of the glide.

I've done the opposite, adding E-On-Coasting to a long habit of DFCO ;)

As you're not using any fuel anymore, both should see "infinite" mileage ...
It's got to be a software thing.


Driving my sister's Skoda Fabia 1.4 Diesel, I've noticed its fuel consumption indicator switches to the hourly fuel consumption when you're stopped.
(I got 3,6 L/100 km or 65.3 mpg BTW, without even trying hard)

When you say EOC, do you mean Engine On Coast or Engine Off Coast?

Ya, I've got to guess that you mean Engine On Coast, since Engine Off Coast would mean no fuel going through the injectors and therefore infinite (9999 and then some) miles per gallon.

With Engine On Coast, you're still using some fuel to keep the engine idling while coasting. Otherwise (obviously) your engine will stop running. So the mpg's can be really really high, potentially even so high that the scangauge will show 9999 mpg, but never infinite.

i use the coast from circa 60mph to 35ish (over about 400m)then engage 3rd and dfco down to 15 selecting 2nd and ready to stop or accel if the give way is clear:-)

Good luck with the coasting!

Sorry about that, I meant Enine OFF Coasting, they do not read the same, or at least not the same on my Scangauge.

Re. coasting downhill in gear to control speed - with injectors off.

IF your car burns oil in any significant amount, you might not want to do this a lot. The high cylinder vacuum will likely suck in a lot of oil - either through the rings or the valve seals, where it comes through to let it burn oil. You'll have a batch of oil in the cylinders when you fire it up with fuel. If so, you'd just be causing it to burn oil

You should be able to test for this. Try it, with medium-high engine rpms, on a longish hill. Restart it when needed, give it maybe 2/3 throttle, and watch for a blue-white oil cloud from the tailpipe. If you see it, it's from the slug of oil built up in the cylinders in that long coast with high vacuum.

If you are reading ~2800 mpg with Engine Off Coast (as you mention previously), then I can only guess that you've got a software problem, as euromodder said:

When your engine is truly off, you are using 0 gallons fuel (0.0000 etc forever), so your miles per gallon would be some number greater than 0 miles divided by 0, which is infinite. There's nothing else it could be, to the best of my knowledge.

Would it burn any more oil in DFCO than not? Seems that it'd be the same amount of oil getting sucked in whether fuel is coming in or not... the valves still open the same...

Valves open the same, yes, but the throttle plate is closed, causing the engine to build a significant amount of vacuum (more than any other normal condition). The effort of sucking air past a closed throttle plate is what provides the engine braking effect during DFCO.

I thought so too, but then I hit the fuel kill toggle switch for DFCO on demand. When I open the throttle plate, all that changes is the exhaust note, not the rate of deceleration.

Combustion pressure casues the rings to seal the cylinder walls more closely, because the pressure increases the force applied to the cylinder walls by the rings.

This is the same reason you will see a certain amount of blue smoke between gear changes in an engine with significant wear.

DFCO with your foot off the throttle creates highest engine vacuum readings, which combined with no combustion will cause higher amounts of oil to be sucked past the rings and valve seals, as was previously mentioned.

regards
Mech

RobertSmalls - With smaller engines, the effect is less significant, and friction/windage, etc contribute a greater proportion of the engine braking effect. In my Jeep, it doesn't get a ton of braking on downshifts at first. Then, after 1 - 2 seconds when DFCO cuts in and the IAC closes, it's like someone slammed on the brakes (big engine gives lots of engine braking).

It also depends on the cam setup, etc of the engine. Some simply provide more braking than others. For example, the engine braking in my sister's 94 Camry 4 cyl is pretty strong, but dad's 05 Camry V6 has much weaker engine braking, even though the engine is bigger.

Hello -

Is "lots of DFCO" a good place for a PCV oil catch can thingy? Or will it "not help much"?

CarloSW2

With regards to my scangauge not reading 9999 when Engine Off Coasting;

As mentiuoned earlier, it is likely a problem with my scangauge ... After reading through this I changed one of my settings to read GPH and when the car is off and the key is in the ON position I still reads .02 GPH usage instead of 0.00 when the car engages the DFCO while cruising in gear ...

Oh well, at least I found my problem ...

Thanks guys ...

Maybe if you set the cutoff value a bit higher it'll work ?

trikkonceptz -

Can you do me a favor and tell me what your RPM and MAP are when the car is off and the key is in the ON position? The reason I ask is that when my key is in the ON position, I still see 61 RPM. That leads me to believe that the SG thinks the engine is still running, however slightly. The 61 RPM could be an error in the sensor at 0 RPM (likely), or some kind of failsafe non-zero value reported by the ECU/PCM to avoid divide-by-zero errors (not likely, otherwise why not 1 RPM for that case).

CarloSW2

Will do, I will try it on my way home and I will post my results as soon as I can ...

I would talk to Linear Logic about your display errors. It sounds to me like a problem with the conversions of the raw data into decimal values.

When I tune an ECM I normally program in aggressive DFCO values as possible. DFCO does create a lot of engine braking, though it is possible to reduce this down to a minimum by setting the DFCO IAC steps to the maximum value of 255. This reduces the engine vacuum during coast-down DECEL, but be aware that if you have a manifold pressure sensor it is often used as one of the threshold parameters for the DFCO logic routine. That means that if you reduce engine vacuum too much after DFCO is initiated, then it will just kick back out of DFCO. There is a hysteresis that will keep DFCO from pulsing on and off.
Also, there is a parameter to unlock the TCC when DFCO is activated, which will reduce the amount of time that you can keep DFCO on. I like to keep the TCC locked and the lower limit TCC threshold as low as possible to prevent unlocking during coast-down.

I always run filters on my PCV system, but during DFCO most of the oiling will be coming from the intake valve guides and past the piston rings.

I will try to post some of the calibration and code for DFCO that is used in my 1991 Chevy truck.

DFCO strategy
 

This is about 95% of the calibration and algorithm concerning DFCO in one particular ECM.

OK as requested I checked my RPM and MAP data while Engine Off Coasting.

My RPM's drop to 0 once the engine shuts off, as far as the MAP reading .... well that must not report through my OBT port because it remained blank with no reading whether the car is running or OFF. So sorry about that one ..


The only one that still puzzles me is the GPM that reads .02 while the car is off and key is on. I have to assume that is the only thing keeping the mileage from going to 9999 while EOCing.

Good luck with your quirky SG ..lol

Here is a screen-shot of some of the (stock) DFCO calibrations on my 1991 Chevy Truck.
link for larger photo:
http://i702.photobucket.com/albums/w...FCOscreen1.jpg

http://i702.photobucket.com/albums/w...FCOscreen1.jpg

Note that the factory calibrations shown are not very aggressive at all, and during normal driving conditions it is rare for DFCO to engage.

Also, DFCO and DECEL-enleanment are two completely different things.
DECEL-enleanment commands leaner than stoich AFRs during most deceleration conditions, but only functions when in open-loop mode.

Though it is very wasteful of fuel, closed-loop mode maintains a 14.6:1 AFR during deceleration conditions unless DFCO is engaged.

no map sensor ?
 

...."OK as requested I checked my RPM and MAP data while Engine Off Coasting.

My RPM's drop to 0 once the engine shuts off, as far as the MAP reading .... well that must not report through my OBT port because it remained blank with no reading whether the car is running or OFF. So sorry about that one .."....

the car may not be a speed density system
it may use a MAF sensor as the primary load sensor
in which case
it would not report MAP value in
Generic OBD2 which is all you get with Scangauge2