I was just wondering if anyone knows how the trip meter calculates the MPG. I had a '86 Saab that calculated the MPG, but it averaged out the last 10 or 20 miles you drove. I am planing on leaving my HCHs second trip meter for an overall mpg. If it only calculates the last 500 miles though, then I guess that would be pointless. If anybody knows how it is calculated, your knowledge would be appreciated.

I think when it wraps it still keeps the mpg reading. I don't know how honda calculates it but it seems to be at less than 1 minute intervals.

My hydroponics experiment

You ever notice how hard it is to lip read cartoon characters?
"Crazy is what the sane call Delta Flyer"

i've noticed that the average mpg meter only hits certain numbers at irregular intervals, for example,

45.0
45.4
45.7
46.1
46.5
46.9
47.3
47.6
48.0
48.4
48.8
49.2
49.5
49.9
50.3
50.7
51.0

the pattern shows 8 steps between each 3.0-step whole number (45-48, 48-51). not sure why. my guess is that the mpg calculating algorithms calculate to the (8/3) mpg unit, and whenever current mpg deviates to a certain threshold from projected mpg from the last reading, the numbers change one unit (or two, depending on the difference)

this involves some simple calculus - if f(x) is the graph of projected mpg, and g(x) is the graph of current mpg, the integral of f(x) from a to b seconds minus the integral of g(x) from a to b seconds, absolute value, is greater than a set threshold (8/3 mpg), then change the numbers up or down (8/3) accordingly. real integrals are approximated by calculators using Riemann sums, which are memory efficient and practically accurate (who cares if the meter is off by 0.0001 mpg?). read up on differential calculus if you want to know more

A tip on the update interval:

On the Insight, pressing the FCD button to cycle through the displays will update your mpg's immediately.

I don't know about the Civic. ??

Regards,

i'm going to have to try this tomorrow

I'm not sure of the formulas, but it calculates the fuel delivered at the injector (pulses) and distance traveled.

I like to leave B as my tank average and use A on individual trips.

Efficient drivers do it better.
1003 miles a tank personal record. 74MPG calculated. HCH1 CVT

It seems the cars milage calcu;ations would be right on the money. Yet when I fill my gas tank I'm always showing a 10 percent or so discrepency based on the fuel put in vs miles driven. I would think that this would even out over time but it hasn't. Its very consistant. Any idea why that would happen?

My hydroponics experiment

You ever notice how hard it is to lip read cartoon characters?
"Crazy is what the sane call Delta Flyer"

I love Hondas. I imagine I'll be driving their cars till I die, but this issue really bugs me. If I could change anything on this car, I'd give it an accurate mileage gauge. How hard can this be? Can't they just put a meter on the fuel line to gauge exactly how much flows through?

I've only owned my car since early January, but I drive quite a bit, and the computer is always at least 3 mpg higher than my calculations. This doesn't engender good feelings. What it does is bring out my cynical side which wonders if this is deliberate. I wonder if Honda wants its customers to feel just that much better about paying a couple extra G's for this car. Honda is so exact and so deliberate in their design in every aspect (this is what makes them great); I find it hard to believe these calculations could be accidentally crappy. Yes, the people who frequent this board will do their own calculations and discover the discrepancy, but the vast majority of HCH owners will not. I mean, I'll be honest with myself, I'm extremely anal about these things, but I'm not so crazy as to believe other "normal" people would or should feel the same way. If I started talking about this with some dude at a bar, he'd tell me I'm nuts. Who cares if the computer is always 3 mpg high? If it says 51 and I really get 48, I should be happy I get 48, right? Not if I'm insane. No, the insane me just wants an honest computer.

If the computer's main purpose is to manage the fuel system, giving me the best mileage possible, balanced with performance, then my question is this: How can it possibly perform this function adequately if it can't even calculate mileage accurately?

I think it's time for a firmware upgrade.

Mine has averaged only a difference of 1.43 MPG between the FCD and my "old fashioned" manual calculations for the 11,000 or so miles I have had the car.

That's pretty darn close in my particular case - they would be hard pressed to improve that with a firmware upgrade !!!

Mine is consistantly 10% optimistic. When I was getting lower mpg it was 10% higher than my calculations. And now that i'm getting higher mpg It is still off by 10%.

My hydroponics experiment

You ever notice how hard it is to lip read cartoon characters?
"Crazy is what the sane call Delta Flyer"