I know theres been alot of discussion on this lately and alot of guys have put alot of thought into what they have posted, but here's a point I'd like to make. If you average your mileage by hand over as few as five fill ups, you can be as accurate as 98% regardless of how full you fill your tank. your error will represent the ammount of fuel that was in your tank on your first fill-up and will decrease each time you fuel up, but will NEVER completely go away.

Here is an example to play with (Kinda long but I cant think of a better way to explain it):

Bob has a Ford Truck with a 10 gallon tank that gets 10 MPG(he don't know 10MPG yet). At his first refuel he had 1 gallon in the tank(he don't know that either) and he pumps 8 gallons thinking his tank is full(it's not, only 8gal pumped and 1 that was there allready). He drives 70 miles. pulling up to the pump for his second stop he figures he just got 8.75 MPG on this tank, also his tank now has 2 gal in it (9gal-70mi at10MPG).

Following this pattern, let me start over and switch to "shorthand"

STOP1--1gal in tank--pumps8gal--Actual in tank 9--drives70mi--by his math 8.75MPG.
STOP2--2gal in tank--pumps6gal--Actual in tank 8--drives75mi--by his math 12.5MPG.
STOP3--.5gal in tank--pumps8.5gal--Actual in tank 9--drives75mi--by his math 8.82MPG.
STOP4--1.5gal in tank--pumps6.0gal--Actual in tank7.5--drives70mi--by his math 11.66MPG.
STOP5--.5gal in tank--pumps8gal--Actual in tank8.5--drives 75 miles--by his math 9.37MPG.

So Bob would say his truck gets between 8.75 and 12.5 depending on whatever. but add all his mpg figures together and divide by 5 fuel stops and he figures he got 10.2 MPG! Not a bad estimate considering his tank was never full or empty. Fiddle with these numbers all you want, change actual MPG at the beginning, change miles driven, change tank size, or use your own numbers from your stops just dont let Bob go more miles than he has gas for or put more fuel in than he has room for and you will see it works out very close-everytime.

One more observation. Technically, no one can say their truck get xx.x MPG, it is constantly changing to some degree. That being said, my example is Technically invalid-in that the example truck gets 10MPG at the beginning. It was necessary to have a fixed number to compare bob's figures to.

 

You sir, are 100% correct. But you really haven't addressed Bob's math capability. And, even if Bob can put the correct data into a four function calculator, he wants to compare his MPG with everybody else. We can only hope everybody else has Bob's math skills.

Once everyone is on the same accurate MPG calculation page, they will want to compare MPG. Very little can be drawn from those comparisons because every truck, driver, load, location, route and weather condition will be different.

Good luck.

 

Anybody who is concerned with fuel mileage readings should do a search on "fuel tank harpooning" Do the mod and it will make life more pleasant.

 

+1 .

 

The only thing you need to add is taking away variables, Like Head Wind, Tail Wind, Terrain, Fuel Quality and a lot of other variables. The only way to do that is longer test period. When I state my MPG figures it over a 5,000 mile period. Anything shorter then that is useless.

 

All these variables clearly indicate that a large data set is best. However, that sure makes it hard to calculate the city driving and highway driving separately!

 

You got me there bismic! You would need 5 fill ups with ONLY city driving to get down to a 2% error and five more for highway.

 

It's in there already Moby. If he has favorable wind conditions for example, he will go farther on that fiil-up. If he drives on four allmost flat tires for a couple hundred miles he will travel a shorter distance before hitting E on his fuel gauge.

TO put it another way: The equation is MPG= Miles travelled divided by Gallons of fuel burned. No more figuring to do- it really is THAT simple. Ther is no need to add or take away the force exerted on a large metallic object travelling throught the earths lines of magnetic flux... or anything else.

I do agree there is a problem with this equation because we can see Bob's math showed different MPG on every tank. We knew ahead of time he was reailly getting 10MPG but he never figured anywhere near this number on a tank. The problem is the figure for Gallons and that is the only problem. He gets an accurate measurement at the pump everyime he fuels but he doesn't have a accurate value for where he started and he can't fill to exactly the same spot every time so it will almost certainly never be right.

That brings us to the main point of my origonal post. when you average just five datasets over 365 miles the error has already become quite low and will get more accurate with each fill up. My previously post included all the numbers to support this. This is much less data than I would have guessed would be necessary, I found it interesting and thought I'd share.
If you require more than 2% accuracy, a longer test period is definately in order, but to say going 5000 miles somehow compensates for the variables you listed doesn't make any sense to me.

 

You are describing an AVERAGE fuel economy. Most people want to know what the fuel economy is on the highway without a huge headwind, bad tires, long uphill climbs, bad tires, etc. That way they can have a basis that is free from these effects. This basis can help you troubleshoot combustion related problems in the future.

Same type of thing in the city except wind and perhaps elevation changes are not quite as significant.

Once you know the fuel economy in these "ideal" conditions, you can apply some factors for adverse conditions like wind, etc.

Anyway that is what I do and how I understand it. I want a baseline that will not vary by 2 mpg. I like the harpoon mod so I ALWAYS know I have a full tank after filling up.

 

I got 1452 miles on my last tank... someone do that math would ya.

 

I don't think you have a camera w/ a wide angle lense that can give us a pic of that monster ...............

 

AVERAGE fuel mileage isn't useful for troubleshooting purposes? I'm just trying to say it's quick and easy to figure and pretty dang accurate (2% error on 10 MPG is only 0.2 off) while trying to figure exclusively for Hwy or City seems a little like trying to nail jello to a tree. I haven't checked the harpoon mod yet though.

 

Well if you put the whole 91 gallons in you got15.956043396604396 I didn't no how close you actually wanted to calculate it.

 

Not for me anyway. The variability in driving conditions is EASILY 3-4 mpg. I want to know when my "baseline" mpg (ie without external factors helping or hurting) drops by 1-2 mpg.

 

man I still have to take pics of that.. man im lazy