Is it just me? At every fill up I'm putting in or three more gallons than what the trip display says I use. I can't help but to think that this impacts the otherwise stellar *reported* mileage.

 

I did my first fill, about a half a tank, on Saturday. It was within one tenth of a gallon of what the truck reported.

 

I've tracked every drop of fuel that has gone into the truck and compared with Trip B which I have never reset. The fuel consumed according to the computer and my log file are within 2% of each other. Call that 98 percent accurate or 2 percent wrong. If you're off by several gallons each tank, your fuel flow sensor has a problem.

 

From my understanding, the lie-o-meter only counts fuel burned while driving. Idle fuel is not counted in the fuel used/mpg etc displayed. In other words, the difference in what the computer says you used vs the actual amount you put in is how much fuel you burned while idling.

To some degree the MPG displayed is correct in the you drove X miles and used X fuel to drive those miles. What it doesn't take into account is the fuel you used to warm up the engine, etc to be able to drive those miles in the first place. I believe this to be a main factor in why so many people complain the computer is wrong when they hand calculate their mileage. Hand calculated is counting all gallons used, driving or idling. Computer isn't counting idle fuel.

FWIW, I'm usually within 1 gallon or so of what the computer says I used vs actual gallons put in.

 

^Bingo. We have a winner. I have excessive idle times > 25%.

Thank you gents. I've been all over this site and I've never met so many with such good information. I wish I could meet you all.

 

I idled for 6 hours and the gallons used showed up on my display.

 

I idle a fair amount in the winter as well and still my LOM accuracy is high.

Come to think of it, just where would Ford put a fuel flow sensor? It can't be a simple meter sitting in the fuel stream. With a recirculating system, you have way more fuel flowing back to the tank than is actually burned. How would a common rail fuel system be able to measure fuel actually injected? You'd have to have a meter in every single injector. There's more to this than I was thinking.

 

My understanding of how the LOM works is the computer calculates fuel used based upon number of injection pulses and pulse width. The pulse width and number of pulses varies according to throttle position, speed, barometric pressure, temperature, etc., all calculated by the ECM.
For whatever reason, mine is alway underestimating fuel used by about 5%.
Epic, is your 2% evenly distributed about the mean?

 

Not mine. If I sit idling, especially if I have reset recently, I can watch the mpg go down before my eyes...

 

I've seen that happen on mine too.

 

Epic, another way you could do it would be to put a fuel flow sensor on the supply line and another sensor on the return line. Then you just subtract the difference .

I'm not sure, but I don't think our trucks have an actual "fuel flow sensor" per se. Please correct me if I'm wrong. The way it's typically done is by calculating fuel used by aggregate injector pulse width and fuel rail pressure.

-los341

 

Hey, those fancy math terms are lost on me...

Mine is a simple average over time since inception. This chart shows summary data including total gallons pumped compared to total gallons used as reported by trip B (1,589.4). Apologies, error is 2.61%, not 2%.



I've also tracked average economy per tank by simply dividing miles traveled by fuel pumped. This graph shows the ups and downs of per tank values and includes a price per gallon line in red. Note the price of fuel slowly climbing over the past 20 months. You can easily see the towing miles, urban miles and highway mile spikes in both directions.

 

I have tracked every fill up since it was new using a product called fuelly.com. They have an iPhone application that makes it really convenient. My computer is usually within 1 MPG of my hand / fuelly calculation. Quite often it is within .5 MPG.

 

If that's the way it's done, it's quite a feat of engineering. You'd be talking about millions of tiny data points in a very short time. The rounding on each individual point would add up to something. Count me impressed if this is the method.

 

So, I'm taking a WAG and assuming your calculation for LOM Accuracy is Actual MPG / Indicated LOM MPG. If this is the case, you're looking at an average 2.61% overestimation by the LOM of your MPG.
I've noticed that most manufacturers LOMs tend to over-estimate MPG, I'm guessing because it makes drivers feel good thinking that they're getting better MPG than they actually are. Anyway, that's my conspiracy theory.