Unfortunately, gpsman1 is so focused on proving himself right and declaring someone else wrong that he ignores the math involved.

The MTE is a multiplication of an MPG times the output of the A to D convertor readout that provides the fuel level in the tank.

So, one would reasonably assume that the MTE will continually change as the fuel level changes. One doesn't have to be a rocket scientist to recognize this.

Therefore if the MPG changes only slowly, aside from the actual level change induced from consumption of fuel, the MTE will change only slowly.

One other thing I observed is that the MTE doesn't necessarily change mile for mile with the odometer. It appears to be more closely related to instantaneous MPG, but I need to watch this more.

Throughout my posts I reported what I observed on my 06 FEH readouts without attempting to cajole people into believing a given perception of how Ford implemented their software. I figured the math would be easily recognizable.

I learned years ago that people have a tendency to believe a digital readout as if it's a rock solid fact. There are many examples of this, perhaps the worst is the long ago KAL 007 flight that strayed over Russia because its navigation computer had incorrect input info entered producing a course to fly that was wrong... and the then Soviet AF shot them down when they entered Soviet airspace.

The pilots blindly followed the navigation readout with out running the result through the common sense test.

As a flight instructor (CFI, CFII, MEI) I learned to teach student pilots to "get their answer in mind first" and then compare with the information in front of them. If they aren't close then something's wrong. Go root it out and make sure you understand the reason for the difference.

Another aviation example was the Air Florida flight that flew into the Potomac River drowning people. The pilots improperly believed the engine pressure ratio gauge when a simple common sense check of the position of the throttle handles would have told someone that the engines were at half power despite the instrument reading. The EPR gauges read incorrectly because of icing on the turbine inlets and they couldn't get the plane out of ground effect at half power.

Get off your high horse gpsman1.

 

So it does not use an estimate of fuel remaining based on fuel flow to the engine? Wouldn't that be more accurate? I didn't think the tank sensors would be that sensitive.

Or is the "A to D" converter checking fuel flow? I'm a software engineer, not an automotive engineer...

 

To my knowledge the calc uses the fuel level. It wouldn't surprise me to find out that fuel flow is somehow included but in my observation the things that seem to drive the readout numbers are MPG & Gallons.

If they used fuel flow they would have to include an integrator in the process. Not a big cost driver... but then when you make profits by saving a few pennies here and a few there multiplied by the number of cars sold, why include it?

The numbers were already available more simply, so I inferred that this is probably what they used. Without reading out their software we are all inferring what Ford did when they built the software.

That's why I tend to observe things for awhile before drawing conclusions. All of my previous posts were grounded in things I observed last summer when I towed and hadn't yet figured out there was an MPG reset capability. (Like I said in a previous post - duh-h-h.)

Sometimes I refill the gas tank without turning off the engine. So the fuel level changes to reflect the new full tank level slowly because of the anti-bounce circuitry. The few looks I took at MTE when the gas tank level was rising to the new full level, seem to reflect a slowly rising MTE consistent with the slowly rising fuel tank level. Go take a look and see if that's what you observe.

Best I got. See if it passes your common sense test.

 

Look, all I'm saying is... to EVERYONE.....

That if fuel tank level falls faster then expected, the MTE will and indeed does make rapid "corrections" to account for this, wheather the fast decrease in tank level is from towing ( and using lots of fuel ) or if you have a hole and are losing fuel.

That's it. None of Bill's remedial math class required!

I'm saying what he's saying... ( I Think ) It does not count down one mile of MTE per mile driven... not at all.

When I first start to tow... my long term average MPG is ~35 and my MTE is ~ 480 miles. Then I tow my giant trailer and get 19 MPG from the start.

After 10 miles my MTE might be 470. ( off by 220 ) down 1 per 1 driven
But after 100 miles, my MTE is down to 300. (off by 140) down 1.7 per 1 driven
After 100 more miles, my MTE is down to 100. (off by 40) down 2 per 1 driven
And by 0 MTE, I still have exactly 1 gallon of reserve, and since I reset my MPG meter and I know I'm getting 19 MPG, I know I can drive 19 miles past zero.

What's the big debate here?
If you call that "MTE moving slowly" well......................

 

Experimentation has shown the on-board electronics use 13.7 gallons as a "normal" fill. ( IE level gauge is at max. but we have also learned if you park on a slope you can get more air out and over fill to 17.x gallons, which would, after a while, make the MTE decrease SLOWER than 1.0 per 1 mile driven. )

 

Your fallacy is the "rapid corrections" you speak of. Nothing is corrected within the computer calculating the number.

The only thing going on is that the error decreases as the fuel level falls. That's the "remedial math" that you don't seem to understand.

If the MPG is not close to reality, and it wasn't for my ~2300 miles towing last summer, then the MTE calc is just as wrong.

Just because the value of the error gets smaller doesn't mean any correction is going on. The value of a 10% error necessarily gets smaller as the value of the number it multiplies gets smaller.

That doesn't mean the error percentage is smaller. When you get to zero the value of the error will be zero. Does that make you feel better?

Your use of the word "corrections" causes an inference that is just wrong and will lead some astray.

 

You revoke your use of the term(s) MTE only moves slowly,
( I've watched it move 2 per mile driven, and I've also seen it make 60 mile "jumps" all at once... what I call "corrections" )
Then I'll revoke use of the term "corrections".

Truce?

 

From the beginning I have reported what I observed last summer while towing...

Why do you use words like revoke...?

Those jumps aren't corrections, they're changes in one of the two numbers that make up the answer. That's my objection to your choice of words.

If you create the perception that this is self correcting, this will lead some astray.

Maybe it's of lesser importance in the overpopulated east, but out in the west, where there can be 100 miles between gas stations, a misunderstood 10% error can ruin your whole day.

 

Either you were an English major, (and are very picky) or were the farthest from.
From Merriam Webster:

re-voke (verb): to put an end to; to annul; to take back; recall; to call back

Which word would you have used?
I feel this whole discussion has been about semantics (noun, plural): a)the study of meanings of words or forms, b)the language used to achieve a desired effect on an audience especially through the use of words with novel or dual meanings.

 

Thanks to everyone for the replies and interesting discussion. It seems there is some controversy about how MPG and MTE are calculated. Based on your inputs my observations so far, I think it goes like this:

- There is an "invisible register" in the computer that keeps track of average MPG over the last tank. This register is independent of the AVG MPG readout register we have access to, which can be reset at will. Maybe folks with a scangauge can access this register? The contents of this register are maintained at all times, ie they don't get reset at fill-up.
- Instantaneous mileage is calculated based on speed/odometer and fuel flow readings. This mileage is integrated over time to give the average readout we have access to. The integration period is reset when the display is reset.
- The invisible MPG register is updated over time based on input from the integrated instantaneous MPG readings. The result is a slow-moving change to the register contents. It is not cumulative, but based on a "rolling average" of the mileage over the last full tank. This gives a convenient MPG reference for calculating MTE.
- The MTE calculation is complex and based on multiple inputs. First input is the average MPG in the invisible register. Second input is miles driven based on the odometer. Third input is a time-averaged reading from the fuel gauge, probably only taken during periods of minimal acceleration/deceleration and incline to keep the noise to a minimum. I'm pretty sure that odometer reading is in fact a part of the algorithm because when in EV mode, I see the MTE drop. This would not happen if the calculation was based purely on fuel level and AVG MPG.
- Instantaneous MPG has no input to MTE. It affects the invisible AVG MPG register, which moves very slowly since it is a rolling time averaged quantity, but otherwise has not effect on MTE.

My analysis above, as I said, is based on my observations and on inputs I've read in the replies to this post. I'm sure there are other factors that come into play, but the above is the simplest explanation I can make. Comments welcome!

RayP