Completely unexpected MPG results

[atomicradish]

Originally Posted by 2016 Versa

It may catch up with me tomorrow but I've been doing it to all my cars for years and hundreds of thousands of miles without any damage.

You've just gotten lucky or perhaps it varies by manufacturer. I used to do the same thing without incident on some of the older cars I Ecodrove back in High School & College but that all changed after topping off my Accord one day. I got a great price on gas at a station far from my home so I filled it up as much as I could. Shortly after the car shut down while driving and I got a CEL for the EVAP system. It took several attempts to get the car to start back up and then it shortly shut off again.... in heavy traffic. The issue cleared up, but I learned my lesson. Its a VERY pricey repair totally not worth risking IMO.

Thanks to everyone who chimed in. The station I frequent does have a slight slope so that probably contributed, along with pump inaccuracies. I'm going to assume the SG was probably closer in accuracy than the classic fillup data. I was really hoping the SG would enable me to get a better understanding of how much gas I'm using on a "per shift" basis with food delivery, and to isolate that data from my routine driving to/from my job (whenever that comes back, thanks corona) but it looks difficult to impossible if I need to drive it down close to empty. Oh well. I feel like I kind of wasted money on this device. I already know how to save gas -- keep my foot on the accelerator as little possible, brake as little as possible, and shift into 5th ASAP.

[nemo]

Originally Posted by atomicradish

I feel like I kind of wasted money on this device. I already know how to save gas -- keep my foot on the accelerator as little possible, brake as little as possible, and shift into 5th ASAP.

You may be surprised by what you can learn from it.

[Fat Charlie]

You don't need to go close to empty, it's just the fastest way to get the most accurate calibration. Longer distances beteen inputs make for more accurate results, just like a 36" level is more reliable than one on a keychain, and a 72" level is even better.

You won't get accurate "per shift" data until you're calibrated. If you admit that and go deeper into your tank a few times to get it calibrated, you'll start getting accurate numbers, which will pay off in more ways than simple shift data: a well calibrated gauge will teach you more about your car than a generic "keep my foot on the accelerator as little possible" rule. There's a time and a place for everything, and your car, tires and feet are different from other cars, tires and feet. Once you're calibrated, go ahead and reset the long trip after every shift. Have fun!

But the short trip numbers will teach you more about how to get the most out of your car than any list of short fills can ever hope to.

[MeteorGray]

All good advice.

Trying to fill up the tank after using only a small proportion of its volume will excessively expose the driver to the vagaries of the process: angle of the sitting car, temperature of the fuel, peculiarities of the pump, etc. IOW, if you habitually fill up a 3/4 full tank vs a 1/4 full tank, your error factor from the fill-up process itself is significantly increased.

When I first got my Surge Guard, I followed some sage advice and adjusted for only about half of the differential between the pumped-gallons vs SG-predicted gallons at each fill. This way, it prevented a lot of under-and-over-shooting of the target. Most of my miles are on the highway under repeatable conditions, and my SG routinely gets me to within a half gallon or so at each fill. The error is only a percentage point or three over or under. It's close enough that I can't do any better because of the maddening myriad of things that affect fuel economy at any given time on any given trip. Incidentally, I haven't adjusted the "gallons used" factor after maybe 10 iterations when I first got it. After that, it has been amazingly accurate as set.

The SG has certainly improved my ability to squeeze all the distance I can out of a gallon of RUG (although I don't employ enthusiast extremes such as coasting-with-engine-off, etc). It's even allowed me to manage my all-important speed better because I can calibrate the SG to show my exact speed vs the vague analog speedometer in my particular car, especially when I changed the tire size to a slightly taller aspect.

The SG is not perfect, but it's as good as it gets as far as I can tell.