DFCO questions
 

I just found out (yesterday) that my wife's car has this feature. maybe more cars have it than I know about. her car is pretty far from being fuel efficient.

she drives a honda element with all wheel drive but it does have the 2.4L 4cyl though.

she is getting about 24.5 MPG. I am wondering about how the DFCO works. I have always disregarded any knowledge about the DFCO thing because it never applied to me.

how can she better utilize this function and is it possible for a vehicle like the one described above to achieve 30MPG?

any suggestions would be great. I just recently convinced her to hook up the scangauge in her ride. if nothing else, it is a start.

Your Cavalier doesn't have this function?

I think it was my post from yesterday that was the source of your "Eureka DFCO Moment." The way that I have been using it mainly on approaching red lights. For instance I know the exact spot on the off ramp towards my house to release the throttle at 50mph so I can cruise the rest of the way in third gear to the stop light. Since this off ramp has an uphill curve to it a lot of people end up slowing down at the curve, accelerating after the curve, then braking at the light. Compare that method and gas usage to my DFCO method with ZERO gas usage. Now attempt to do this with every off ramp, traffic light, traffic jam, stop sign, intersection etc. etc. and the MPG figure starts to increase.

No offense to your wife, and no offense to the female hypermilers on this site but what I explained above is way too much hassle for my wife. No matter how much I talk about FE I'd never expect my wife to put any of those techniques into practice. Most FE techniques are way too much hassle for the average male driver too.

I doubt any car from this century is completely unable to DFCO (unless of course it's been modified). I also doubt any Honda after maybe 1995 entirely lacks DFCO.

It seemed like a really great thing when I first learned about it, but its real world usage is limited. Really all you can do is be aware of it and try to optimize for it while slowing/stopping (by knowing the exact conditions in which it works and trying to stay in those for as long as possible), and also try not to feather the throttle -- either apply some power or get completely off so it can DFCO.

the diffucult thing (as stated above) is that it is on my wife's car. she is just now starting to understand that if we can lower our fuel costs then we will have money for other things. she isn't willing to do a lot of stuff but the sheer fact that she has started to care may mean that she is changing her mind about all of this.

also I assume that if I were to buy a new honda or toyota that it will have this feature as well? my car doesn't have it but it is a 97 chevy so I can't expect it to have all the cool stuff anyway.

The wife factor...yeah, that's a tough one. I knew it was a lost cause right from the start and didn't bother. Instead, I mostly just kept it to myself. I occasionally gave her a brief description of a gas saving technique, or talked about it to someone else while she was around, and I drove with some hypermiling practices while she was a passenger but did not say anything unless she asked. She has decided to do a little, and tried P&G which provided a decent FE benefit but isn't compatible with her Isuzu (see my thread titled something like "I've finally found an automatic that doesn't like neutral coasting").

FWIW, my 2002 GMC and my 2008 VW both have it, though it's mostly useless on the GMC due to the delay before it activates (I suspect that a custom tune could change that).

If a 2002 GMC full size V8 pickup has it, and a 2008 VW with awfully short gears/close ratios for boy-racers has it, I would be hard pressed to believe that any new car, Honda, Toyota, GM, Ford, or even Kia would not have DFCO.

I wonder if the Beast has it? I don't think it does. Its the last year of that body and the 350 V-8. They probably started with model year 1999 trucks. Then again, it might and like you said its virtually useless because of the delay.

There's a sure way to find out for yourself. I think you know...you've read enough of my posts. Just click on the DIY fuel rate meter link. ;)

I'd guess that it's pretty unlikely, considering the way my newer generation 2002 model with a much newer engine and computer acts. Then again...
https://www.google.com/search?q=1999+gm+5.7+v8+dfco
https://service.gm.com/gmspo/mode6/pd...t57R_LD_tE.pdf
Considering that GM's own service PDF (hey, cool find!) for the 1999 C/K L31 5.7 says to test the O2 sensor while in DFCO, I guess you've got it.

But mine is a 98 with the 5.7. I've got the SG set to report fuel rate, and I've never seen it go to zero while engine braking, but then since I've gotten the SG I've been shifting into Neutral to gain the biggest instant MPG's possible. How long is the DFCO delay? What are the RPM ranges on your truck? I have a couple of hills I can try to force it into DFCO (if they are long/steep enough)

-Jay

EDIT: I changed the search parameter to 1998 and it appears to have it. I wonder what the conditions are required for it to activate? From your comments it sounds like its only useful for engine braking going down a mountain.

-Jay

Oops, I don't know why I thought yours was a '99, that would be the new body style...

ScanGauge isn't good at detecting DFCO nor can it measure fuel rate. It reports fuel rate by guessing and calculating. Set one of its readings to show Open/Closed Loop; if you see Open Loop while off the gas pedal you're almost certainly in DFCO.

In my 2002 5.3, it takes 6 to 10 seconds of engine braking above 1500 rpm before it will DFCO. For demonstration purposes, take it up to 70mph and drop to 3rd; that should put you in the 2500-3000 rpm range and should last long enough to assure DFCO even if you're not going downhill.

I use DFCO with additional braking by turning on the AC, for free cool air. Saves me having to run AC otherwise, when temps are close to 100 outside. Instead of sitting at the light melting, I use DFCO to delay my arrival and provide a little cool air, just in case I come to complete stop.

regards
gary

I will have to put open/closed loop on there again. I took it off the other day for CPM. I wonder then, if going downhill on I-95 if its DFCO'ing in OverDrive? We do have some really long hills on the highways around here. DFCO would be nice to do on the highway.

-Jay

It could DFCO downhill in OD...you'll find out. :)

I'll look for it now... Before I didn't think it was even a possibility. Now I just have to figure out how to get it into DFCO and keep it there. Maybe I can push my mileage up with this. If I can confirm DFCO then I won't be scared to engine brake either. My guess was that it overreved the engine so more fuel is sent to the injectors. As you mentioned earlier too, I wonder what's involved in reprogramming the computer to go into DFCO earlier...

the way I noticed it on my wife's honda was the fact that the MPG meter hit 9999 so I figured that is what it was. I thought it was just on hondas like the VX or something super fuel efficient and sold that way.

obviously I was wrong

I think it is on most cars, at least it was first noticed by me on early Nissan Z cars with fuel injection, beginning in 1975.

When you take your foot off the gas pedal, before FI you had to add air to the exhaust to burn off the hydrocarbons. With FI you can simply kill all fuel to the injectors.

Put the car in any gear and let off the gas at bout 2500 RPM, see if it kicks back in (fuel delivery) at a lower speed. On my VX with AC it comes back on at BOUT 1300 rpm, and will actually keep the car going at a significant speed.

regards
gary

So, any bets on a 76 Eldorado with a an EFI 500 ci V-8?

I know, I'm pushing it... :)

That would be something to DFCO in a 76 Eldorado though...

If it doesn't have an air pump from the factory, it probably has FI and DFCO. at least since 1968-69.

My 76 Z had no air pump, no EGR, and no Catalyst. Federal version. California was different.

regards
gary

The 76 Eldorado was a really high tech car for the time, despite the fact that it was a land barge. They came with FWD, EFI, and a 500 cubic inch V8.

-Jay

Therein lies the rub. If it's not in DFCO, it's using MORE gas than at idle speed...the air/fuel ratio has to stay the same and at higher RPM it moves more air. SG's fuel rate readout ought to give you an idea of just how much you're looking at there.

https://www.efilive.com -- but it's expensive. BlackBear Performance can probably do it, I think they're using a shop-licensed EFILive. I'll probably go with them eventually.

I tried to DFCO on my way home tonight. The big problem is that the max speed limit between work and home is 45MPH, and I'm only on that road for 1 block. The rest of the way is 35 and 25. I tried to DFCO on the 35 mph stretch... I don't think it happened. On the 25 mph roads there is a relatively steep hill a few hunderd yards long. I tried to keep it in the sweet spot. Whenever it fell to 1600 RPM I'd downshift. Still SG did not report open loop. I guess I can't hit DFCO in my normal driving. I will try to DFCO next time I'm on the highway. If there's a way to make this thing DFCO I will find it. I was going to go to visit my parents tomorrow, but I was planning on taking the Buick as she hasn't been driven in a while and she needs a good highway run. I need to drive it more anyway... I haven't put gas in it since I've joined here... She needs some gaslog data.

-Jay

1. DFCO is a lot harder to acheive in an automatic due to the engine revs being able to slip lower than drive rpms through use of the torque converter.

2. Toyota has been using DFCO in FI cars since at least '85 (in the manual there's a way to test it: basically, warm up the car and disconnect the TPS and leave it at idle, then open the throttle and watch the tach. when it reaches 1800 rpm (And thinks the throttle is closed, air measurement via airflow meter) it'll cut fuel till the rpms drop below 1300) and my POS bargain basement 1987 TBI 4 banger chevy S10 has it too. Since someone pointed it out, I can tell every time it goes into "deceleration lean-mode" (as the shop manual calls it) the S10 goes into DFCO any time it's over 1500 rpm and holds it down to about 1300-1400. it takes about 1-2 seconds to engage.

I thought catalyst was mandated on everything 75 and up.

Ok, I took it on the highway (I-66). Had about a 15 mile run before I had to get off. I tried to DFCO the entire way. Not once did SG report open loop. I figured I had totally hosed the mileage on that tank, and decided to fillup even though I still had 1/2 tank because I spied an Exxon station @ $3.749 which is 20 cents cheaper than the cheapest gas in my town. One strange thing though... SG had overestimated my fuel usage by about a gallon, and on a 10.8 gallon fillup that is a considerable percentage. Now this is what I'm wondering... Was the calibration of my SG off because I was running a high concentration of methanol in my tank when I calibrated it, or was I really DFCO'ing all the way down I-66. Of course as discussed earlier SG has no idea the fuel is cutoff in DFCO, so maybe the fuel was cutoff, but SG was reading the RPM's and assuming I was using more fuel?

Anyway, I'm calling this a rise in mileage even though its technically .3 MPG lower than my last tank as this tank was mostly short trip city driving, and the last tank was mostly light highway driving. I just have to figure out why it went up. Lower methanol concentration in the tank? 20 miles of highway DFCO attempts? Combination of the two? Something else I'm not even considering right now? Everyone is welcome to look at Beast's gaslog and contribute their thoughts.

DFCO Confirmed!!!
 

Well everyone, I did confirm DFCO with my Scangauge! Open/Closed loop does not work, at least not on a 98 GM truck. I did email SG's tech support for help, but I figured something out before I got a response from them. This is what I was thinking... I can't confirm DFCO by fuel rate, open/closed loop, or FI duty cycle with the SG. One thing I could monitor was the O2 sensor output!!! I programmed Bank1, Sensor1's output as a programmable gauge and named it o2. I then setup my bank of 4 gauges... MPG, RPM, TPS, and o2.

I left mom's house after lunch and went on my drive back home. Going down the backroads I had the cruise set on 45. I noticed in driving most often the reading from the O2 sensor would be between 50 and 85. I then got on the Prince William Parkway. I set the cruise at 55, tranny in OD. Going downhill where you would ordinarily gain a bit of speed I noticed something... and I wouldn't have even noticed it except the O2 sensor readings were clueing me into something going on.

Right as I crest the hill and start going down the cruise lets off the throttle. TPS = 0. If TPS = 0 and RPM ≥ 1500 after about 3 seconds I will ever so slightly feel the TC lockup. Right when the TC locks up the O2 reading will jump to 80 - 85 for about 3 seconds, then the O2 will read 0 until either the RPM drops below 1,000 or throttle is applied. When DFCO cancels I will feel a bump from the TC, then the O2 readings will go back to the normal range. At highway speeds of 65 MPH in OD it pretty much DFCO's whenever the cruise lays off the throttle for more than 3 or 4 seconds. This explains why I haven't seen the mileage gains I expected when I take it nice and slow and set the cruise at 55 on the highway. @ 55 in OD its borderline whether or not DFCO will kick in unless you are going down a grade.

This explains why I'm not getting the mileage gains I expected when neutral coasting and driving slower. I ended up disabling DFCO with those other hypermiling techniques. Now that I know what's going on I hope to see a bigger MPG gain on my next tank, and can better judge when to neutral glide, and when to leave it in gear and DFCO... :)

-Jay

I bet you are feeling a bump from the TC unlocking and then locking again when you hit the gas. Those larger vehicles tend to do that to reduce engine braking and increase coasting distance.

My Cressida used to do that but the Tracker doesn't. In fact, I can go into neutral on the freeway and the converter stays locked even when I'm going back into drive.

Ford Man, I think Honda CVCC engines did not have a catalyst. I know for a fact that my Z did not. California versions of the 76 Z did have a catalyst.

The Z even had the large filler neck hole that allowed you to use leaded fuel, even though none was available.

regards
gary

But if the TC is unlocked you won't be able to turn the vehicle's engine with the vehicle's momentum. My theory is that most of that movement would just be swallowed up by the TC, and not turn the engine. What it feels like is if you were driving a 5 speed and you crested the hill with the clutch depressed, and in 5th gear. It feels like the clutch was brought halfway up. Not a big jerk, but once I knew what I was looking for it was easy to notice.

-Jay

Why not? An unlocked TC isn't like neutral transmitting zero power, it just transmits less than full power. It would just turn the engine less fast than if it was locked.

I think your TC isn't toggling, you're just feeling DFCO engage and disengage.

I know for a fact that all computerized Hondas utilize DFCO. The earliest Honda ECU that I have personal knowledge of was a 1988.

DFCO generally requires zero throttle, not merely low throttle.
 

FWIW: One observation I made about DFCO, that hasn't been mentioned yet, is to make sure your throttle really can go all the way down to zero (and not just fairly close to zero). And it might not be possible to have true "zero throttle" if/when the some mechanic (inadvertently) adjusted the throttle cable just a little too tight.

I know, as that appears to have actually happened to me. When I got the CRX, it didn't seem to want to go into DFCO, even after I learned it should be possible. However, I also noticed (independently of the DFCO goals) that the throttle was very "sensitive" and I accelerated at (high) foot angles/positions I didn't like. So to try to fix the foot position "problem", and allow/require me to press a little harder on the foot pedal before getting acceleration, I (found and) adjusted the throttle cable tension "screw" (it wasn't really a true "screw", but you adjust it a little like it was one) to give the throttle cable a little extra slack (maybe all of 2-4 millimeters extra slack, but still I made the cable slightly more "loose"). And that small amount of extra cable slack was all the throttle needed to make my foot angle/position much more comfortable on my foot. And as a side-effect of that cable tension adjustment, I also noticed that the CRX starting easily going into DFCO (when my foot was off the pedal, and my RPMs were up sufficiently), whereas previously it rarely went into DFCO even when I took my foot all the way off the pedal.

i.e. I'm speculating that some former mechanic, in an effort to remove all the "slack" from the throttle cable, actually tightened the cable just a bit "too far", so that what should have been "zero throttle" (because my foot was off the pedal) was actually a very tiny amount of throttle instead. And this small amount of throttle always being present, while not overly harmful to normal car driving, appears to have been just enough gas/throttle to turn off (and keep off) any attempts at DFCO, because ECUs generally require "zero throttle" to kick in DFCO and I was apparently never getting fully "zero throttle" (due to the slightly "too tight" throttle cable).

So I guess the moral of this story, is to check the tension on your throttle cable, and consider playing with that cable tension adjustment if it seems appropriate. Because if/when the cable is even slightly "too tight", it can prevent you from fully going to zero throttle, which will also have the side-effect of preventing DFCO.

Ok, I ventured inside the beltway today... I went to Alexandria. Got pleanty of practice DFCO'ing at traffic lights in a completely urban environment. By carefully and progressively downshifting I can keep DFCO going down to about 5-10 MPH. SG was reporting 18 MPG average on my urban driving with DFCO. Not sure how accurate that's going to be because I've already established that SG doesn't know what's going on with DFCO in my truck. Going 55MPH on urban highways inside the beltway have few opportunities to DFCO because of heavy traffic.

Once I got on the toll road on my way out of town I again set the cruise at 60. I'm thinking 60 is probably the most efficient speed in the truck. I used to think that the slowest speed possible and still be in OD would have been best. I don't think so now. If you touch the pedal, or loose speed it drops down to 3rd gear - something we want to avoid. 55 is good, but its borderline whether or not DFCO will kick in when the cruise lets off the throttle. @ 60 you are going fast enough that you can stay in OD, even if you give it some throttle or loose some speed. Plus 60 is fast enough that if the cruise lets off the throttle for more than 3 seconds it DFCO's. There was one stretch where I was able to maintain DFCO for almost a whole mile.

I'm still wondering if there is some setting in SG to make it realize I'm in DFCO and reflect that in the instantaneous mileage.

one thing I have been told about the scangauge is that it uses air flow to get MPG readings so the engine is still pumping air through it with DFCO. I guess that is the big reason that the scangauge doesn't always see DFCO.

I just use the scangauge as an indication tool anyway. wait until the tank gets filled again, that is where it really matters in the end.

Yes, the SG is a tool to help me. Right now I'm using it in ways I never considered when I bought it a month ago. If you told me a month ago I'd be watching my O2 sensor output, and really caring what the numbers were while I was driving I would have said you were crazy.