Does anybody reading here know how the MPG is computed for the display? It's got be some sort of moving, weighted average; but the basis for it just doesn't seem to be very clear (to me).
The overall trip mileage for a recent 3k+ trip was surprisingly accurate according to backup calculations for that trip. But about a mile into a moderate climb, it dropped several tenths. And that just can't be right given the average mileage I get climbing, and how that number needed to be factored into the overall mileage.
I reset my MPG computer each fill up. I carry 59 gallons, that gives me anywhere from just shy of 1000 miles to just over 400 miles between fill up. I have found that it has been consistently very accurate, never more than
2/10 of a MPG off, and if off, always lower than the actual calculated value (IE computer 15.6 ~ calculated 15.8).
I don't have any technical data to back this up, but with all the sensors in place, engine speed, fuel consumption rate, intake air temperature, turbo boost, travel speed, etc. an accurate MPG calculation should be quite simple.
The computer knows how much fuel is delivered to the cylinders, so calculating how much will be needed per mile is not a difficult task.
The display is an average since the last time the counter was reset. If you reset the counter often, then the changes as yo climb a hill will be great. If you seldom reset the counter, then the change will be very small since repeated activities will constantly be included in the calculations.
My car keeps track of two Consumption Rates. I like to reset one of them every time I get gas, and let the other one keep track of history. Currently, I have 25 mpg on the first counter because it shows recent activity of driving on the freeway a long distance, the other counter shows less than 20 mpg because is includes lots of in-town driving, and driving up the steep hill to get me over the mountains to the coastal areas.
The guage works like a vacuum guage, sorta. Actually, it has nothing at all to do with vacuum, but the way it works reminds you of vacuum. It (the guage) is looking at the fuel injector timing pulse. If the pulse is small, then the fuel delivered is small and the mileage will be large if the speed is high. If you can maintain 70 and use short fuel pulses, then you will have high mileage, but if there is a head wind or a hill that demands a higher fuel pulse, then the mileage will drop as a result. If the hill is nearby to the fueling station and you reset the Consumption calculator, then you will see a large change as you climb the hill. If you let the Consumption calculator run through two or three tanks of fuel, then you will see no change in the consumptin rate as you go up the hill. The reason is, the calculator shows a Historical Average, but if there is a lot of history, it changes very slowly, but if there is no history, it changes quickly.
There is no point is displaying an arbitrary number. The number _should_ be accurate to a .25 mile, or it is useless, and there is little need to expend Engineering Effort to display useless information.
Well, that's kinda what I thought...but it doesn't seem to be working like that. We took a weekend jaunt from Everett to Yuma. ~3500 miles all told. We spent a day in and around San Jose, another couple in and around Yuma, and some other places in between. Most of the trip was >65mph, and a big chunk of that was just over 75mph. I reset the counter before we left Everett. I saw ranges from 7MPG to 22MPG...but things started to settle out to 18.3 as we were nearing Oregon on the return. As we were climbing out of California, it dropped a couple of tenths. Given that the average was being calculated over nearly 3km, I wouldn't have thought that a few minutes into the climb, that I'd drop a couple of tenths.
To be sure, I got them back on the way down the other side. But I just thought that with that kind of history, I should have been on a long, long, hard climb before it had that much impact. That's what made me think that it was some kind of weighted average. As for the system counting the fuel pulses, yep, that's what I'd surmised (cuz I know that FF meters are nowhere near accurate enough).
In any event, ~3200 miles, '06 F350 6L TD LB SRW, I consider 18+, A okay! I was just wondering how the calcs were done. Next week, we'll be doing that again, except that we'll be schlepping a trailer...it ought to be interesting to see the differences in actual practice.
Thanks again, Jeff, for your most informative reply...
Jeff is really close on this one, but it doesnt look at the fuel pulse width, it looks at your throttle pedal, fuel level in the tank (going up a steep climb will change fuel level pretty quick). It will then ESTIMATE your fuel milage at that point. These things arent designed to be that accurate. Someone with a heavy foot, and resets the MPG guage every fill up will see it drop extensively when they take off from a stop. The computer just knows that if you are giving this much throttle at this speed, you should be getting this many MPG with this much fuel usage. You are lucky if yours is that accurate, most arent.. LOL I took several calls this summer dealing with compaints about economy meters and how inaccurate they can be.
Okay. I think I get that. But I was talking about the MPG computation, not the range until empty computation. While the former must of necessity be used to compute the later, the MPG value must be independent of the level of fuel in the tank.
However, given what you just said, I think I'm going to spend some time watching the "Miles to empty" value and see how much it varies as I go up and down hills (that'll give me something to do on those mind-numbing stretches of I-5)...(:-o)!
I can imagine! Although, for me anyway, my MPG display seems to track spot-on with my hand calculations.
What would really be interesting would be to have two displays. One to display the mileage from the last reset, and one to display the instantaneous mileage over some shorter interval...let's say the last
10-miles or so. That way you could use it to help you moderate your foot and work towards a better touch...
My '94 BMW does this too. It gives the instant fuel consumption reading on a guage that moves much the same as a vacuum guage might move -- but, beyond the appearance there is nothing similar to vacuum stuff going on -- and gives the Average Consumption in another place on the instrument display. The Average Consumption is available in two readings, one is long term and the other is short term (or both can be the same, but that defeats the purpose of two readings).
I also get a Range-to-empty display that looks at the remaining fuel load as well as the consumption rate.
MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.