EOC ~ only useful at low speeds
 

The mpg using EoffC on my trips has been much less than with leaving the engine on when coasting. For example, two previous 125 trips averaged (1) 54.0 mpg using EoffC all the way, then (2) 59.9 mpg using EonC, averaging about 55 mph. Care has been taken to shift at lower speeds, and to not lose momentum with shifting, so the continued poor EoffC result has been puzzling. My kill switch is connected to the fuel injectors.

Recently I've noticed that switching the injectors back on (with the engine off), drops .05 volts from the battery, thus 20 switches drops the battery from 12.8 to 11.8 volts. Perhaps this drop results in the alternator working harder while driving, uses more fuel, and causes the lower mpg.

My hypothesis is that EoffC is only useful below a certain break even point, depending on the aerodynamics of the car. For example, my car gets 30% better mileage using EoffC when averaging only 20 mph, but 10% worse mileage averaging 55 mph, with an estimated break even point of 40 mph. Better aerodynamics would raise the break even point higher.

I don't know about the electrical bits there, but at higher speeds your throttle is more open already, so less return from p&g.

That's a 10% difference. What mileage obtains with no p&g? Does the voltage not rise again with the engine back on?

Same speed. Same aerodynamics. Small sample size. Maybe you'd do well to have the mods move this to General Efficiency.

Looking at the glide distance, aero drag is stronger at higher speeds so you'll lose speed faster at higher speeds, which cuts its utility. Looking at fuel consumption while gliding, you'll have more injector pulses per foot travelled at lower speeds, so the benefit of engine off is magnified there.

But I wouldn't say it's only useful at lower speeds, just that at higher speeds you need a longer glide for it to make sense, generally with some sort of a downhill to help extend the glide while minimizing speed loss. It doesn't take much of a downhill to help- one of my morning EOCs is on a stretch of the highway that I hadn't realized was an incline until I started hypermiling.

But yeah, higher speed reduces your ability to save fuel.

Isn't that the truth?? Wow, I was clueless to some little gold mines. AS to the OP, Just like he said: higher speeds will largely negate anything you can do, besides slowing down.

How many RPM's at 55MPH?

aerodynamic break point
 

EonC ~ coasting in neutral ~ engine on
20 mph ~ 62 mpg (7 stop signs in 2.7 miles or else would be higher)
55 mph ~ 60 mpg (one stop sign in 125 miles)

EoffC ~ coasting in neutral ~ engine off ~ no throttle
20 mph ~ 82 mpg ~ 33% increase
55 mph ~ 54 mpg ~ 10% decrease

Estimated aerodynamic break point ~ where both will be the same ~ 40 mph

Being conservative, estimating 250 pulses in 125 miles, and EoffC used only 40% of the time, that's 100 pulses, times .05 volts would be 5 volts of drain without the alternator charging the battery. I'm not sure exactly how alternators work, but such increased drag and drain seems likely to cause the alternator to run more.

Rpms are perhaps 900 rpm with the engine on in neutral, and 0 with it off.

That is where you lose me, are you bumpstarting? Is your ecu going into a weird mode (would a cam or crank signal interupt prevent it)?

It shoud always use less power when off.

RPM's in 5th gear at 55MPH?

Yes, bumpstarting.

The battery use is much higher.

Rpms are not relevant here, as the pulses are the same, but somewhere around 2500 rpm.

Yah, but power demands should be lower since you aren't running injectors or spinning alternators and stuff. Though there is probably a break-even point for getting everything stopped and started rotating again. I do wonder if cutting an injector is the best approach though, the ECU is bound to see a fault and may go open loop after restarting, whereas if you switch off the rpm signals to the ecu, it may simply "think" the car is stopped (and you don't need much of a switch to break a signal). How you kill it might be a factor. My honda gets a CEL when I break an injector lead.

I never saw such a problem with my 96. I was keying off momentarily to kill it. Do you have an injector or ignition kill instead? Try using the key for a while and see if you have different results. That battery usage seems more than excessive. I would see a spike in power consumption when I key on, as the fuel pump primes, but then it drops back down.

ignition switch ~ no voltage drop
 

I'm using a main relay fuel injector switch. Turning the ignition (not the engine) off and on 10 times, with 3+ second pauses, only drops the battery by .02 volts total. The key must be turning the injectors on and off, and there is the same thumping sound, so I wonder what the difference is.

fuel injector wire
 

The PGM-F1 main relay ground controls the fuel injectors, fuel pump, o2 sensors, alternator, vehicle speed sensor, and maybe the PCM, one or more of which must be causing the drain.

I've not found a fuel injector fuse, so I'm going to splice into the fuel injector wire, before it gets to the PGM-F1 main relay.

I would tend to agree with John's hypothesis. The more you EOC the more voltage you loose, and that energy then has to come from the alternator. It takes around 7hp for a small car to cruise and an 80A alternator can output 1.5hp, if the alternator has a lot to do you can see that can be a big hit to economy.

When UFI got a battery twice as big as stock, my MPG went up by as much as 15%, I charge it at home and the alt has nothing to do for at least the first 140 miles.

If you've got a lightweight car with poor aero, I've found EOC to be ineffective.

If you've got a 2 ton van EOC works at freeway speeds. I've adding a solar panel to my 1.5 tonne van roof next week to offset the EOC battery drain. Personally, I expect significant gains. Right now I often EOnC out of kindness to the battery when the voltage is reading a bit low already.

Yes. I believe it. I have to do a lot of voltage management to keep things healthy, including overnight charging - daily in winter with headlights but less in summer.

I disagree completely.

larger battery
 

Great ideas! I'll keep those mods in mind. Thanks for the feedback.

I should have added 'at freeway speeds'. Keep in mind your Fit is both heavy and aerodynamic compared to my Fiat 500. Even with my Renault Kangoo (Van) which only weighs 1270kg for a pretty big vehicle (0.33cd) EOC is worthwhile.

I'm right into EOC but there are some cars for which it doesn't work.

Actually, no. Our cars are about as close a match as you can find from different makers. According to Edmunds.com, your Fiat is just barely lighter and about equally aerodynamic. (2363 lb vs 2490, cd 0.33 vs my 0.34). They're both tallish hatchbacks, with the drag that comes with that basic form. The Fit is also slightly bigger, making the cdA worse (66.7 in wide x 60 in high vs your 64.1 wide x 59.8 high).

It could be that your engine management runs rich for a while after a restart or something like that, making the shorter eoc cycle at highway speed less beneficial.

What is your highway gear ratio like? I'm all the way up at 2850 rpm @ 60 mph, so anything I can do to minimize that, like p&g, is a huge benefit. A lower cruising gear would close the gap and reduce the possible gains.

Edmonds is wrong on the CD, it's actually 0.367 and mine is European spec so only weighs a fraction over 2000lbs (and mine's been lightened). I also have the bigger 'Abarth' rear spoiler and 16" wheels so I wouldn't be surprised if my actual CD was around 0.38.

The engine has factory stop start so stays closed loop on restart. It spins at about the same revs as your Fit at 60mph but it's a 2cyl without throttle plate so at least the pumping losses are low, it's also got Fiat's fancy Mulitair valvetrain.

It's a turbo car and EOC is a bad idea for it anyway, but EOnC has shown significantly worse results than DWL (at 875cc it burns very little at idle). Aggressive EOC might brake even with DWL but wouldn't be worth the bother. It takes absolutely forever to get back up so speed an no time at all to scrub it.

DING DING DING DING! We have a winner!

When you power back up, the ECU is programmed to wait for the O2 sensors to warm up. Newer cars have some logic to know how hot they are, but older ones just had fixed timers. 20 or 30 seconds of open loop could kill any savings you got shutting off the engine.

fuel injector wire kill switch
 

I removed the pgm-f1 kill switch yesterday, hooked it up to the main injector wire today, and it works! I've not gone for a test drive yet, but turned the switch off and on 10 times, with 3 second delays, and there was a negligible change in the voltage. Success!

Hopefully this will raise the break even point enough, that EoffC can be useful on trips. However, I think the aerodynamics will have to improve quite a bit, for that to happen.

Does an injector kill switch get around this? Even my 2014 car still goes open loop for about 5 seconds after key off.

short trip 77.1 mpg
 

My 2.7 mile route (backing out of garage to street, 7 stops, back into garage) with the
pgm-f1 kill switch was 63.3 mpg cold. This morning with the new injector wire kill switch
I got 77.1 mpg cold. Eliminating the .05v drain appears to make a difference on short trips.