How to compute mileage (mpg)?

On Mon, 27 Jun 2005 10:32:04 -0700, "Jeff Strickland"
> wrote:
>"Kathy and Erich Coiner" > wrote in message
>news:byhve.1095$4M1.107@trnddc07...

>> 4. Fill tank again. Use same pump at same station. Let the auto shutoff
>> system work. (Do this on initial fillup too. Record gallons to fill
>
>You make this much more difficult than it needs to be. All that is needed is
>to fill the tank where ever it is convenient to get gas.

Well, it /is/ slightly overkill - but if you want a little more
accuracy on your fill (let's say you are doing an economy run test for
a magazine or newspaper article, or just want to get more accurate
numbers on each fill) it's a simple and logical way to get it.

Using the same gasoline pump means you will use the same fill
nozzle, and are parked at the same level (or not level) attitude on
the pavement. Oh, and you want to use the same fill speed setting on
the nozzle, usually the slowest speed.

Gasoline nozzles will vary a bit between different ones exactly when
they trip the automatic shutoff, but the same one will probably kick
within a quart from fill to fill. This will get you a slightly more
accurate fuel mileage reading than using any old pump at any old
station.

The only way to get better numbers would be to install a temperature
compensated calibrated fuel tank with a fill view port on the car,
like the ones your local County Weights and Measures guy uses to check
the calibration of the gas pumps. (Megabucks, and a waste of effort.)

--<< Bruce >>--
--
Bruce L. Bergman, Woodland Hills (Los Angeles) CA - Desktop
Electrician for Westend Electric - CA726700
5737 Kanan Rd. #359, Agoura CA 91301 (818) 889-9545
Spamtrapped address: Remove the python and the invalid, and use a net.

No Spam wrote:

> Amazing. Our educational system on display.

Moves lips when reading stop signs?

Newbie wrote:
> It may seem like a superfluous question, but what is the best way of
> calculating what mmg yuor vehicle is getting? I have my miles reading
> of course, and I can record how much gas I put in, but there was some
> gas already there in the tank, and how do I know what was traveled on
> how much gas? If you have a good system, please share!

It's a basic math exercise usually found in the form of word problems.
For the results to have any value you have to fill up every time you do
the calculation.

Number of miles driven divided by number of gallons equals the miles
per gallon.

A cumulative average can be computed by using the same numbers as above
but just keep adding them in and redividing. After a while the number
becomes glacially slow in responding to change and a performance
problem may not be noticed for several tanks.

A moving average covering some moderate time period like five fillups
is probably the most useful because it retains some responsiveness to
short term changes while also reflecting performance over more than one
fillup. Unfortunately it also requires the most work.

"Bruce L. Bergman" > wrote in message
...
> On Mon, 27 Jun 2005 10:32:04 -0700, "Jeff Strickland"
> > wrote:
> >"Kathy and Erich Coiner" > wrote in message
> >news:byhve.1095$4M1.107@trnddc07...
>
> >> 4. Fill tank again. Use same pump at same station. Let the auto shutoff
> >> system work. (Do this on initial fillup too. Record gallons to fill
> >
> >You make this much more difficult than it needs to be. All that is needed
is
> >to fill the tank where ever it is convenient to get gas.
>
> Well, it /is/ slightly overkill - but if you want a little more
> accuracy on your fill (let's say you are doing an economy run test for
> a magazine or newspaper article, or just want to get more accurate
> numbers on each fill) it's a simple and logical way to get it.
>
> Using the same gasoline pump means you will use the same fill
> nozzle, and are parked at the same level (or not level) attitude on
> the pavement. Oh, and you want to use the same fill speed setting on
> the nozzle, usually the slowest speed.
>

A variation of a half gallon or so on 300 miles isn't going to have a
significant effect on the mileage.




> Gasoline nozzles will vary a bit between different ones exactly when
> they trip the automatic shutoff, but the same one will probably kick
> within a quart from fill to fill. This will get you a slightly more
> accurate fuel mileage reading than using any old pump at any old
> station.
>
> The only way to get better numbers would be to install a temperature
> compensated calibrated fuel tank with a fill view port on the car,
> like the ones your local County Weights and Measures guy uses to check
> the calibration of the gas pumps. (Megabucks, and a waste of effort.)
>
> --<< Bruce >>--
> --
> Bruce L. Bergman, Woodland Hills (Los Angeles) CA - Desktop
> Electrician for Westend Electric - CA726700
> 5737 Kanan Rd. #359, Agoura CA 91301 (818) 889-9545
> Spamtrapped address: Remove the python and the invalid, and use a net.

Jeff Strickland wrote:
> Bruce L. Bergman wrote:

> > Using the same gasoline pump means you will use the same fill
> > nozzle, and are parked at the same level (or not level) attitude on
> > the pavement. Oh, and you want to use the same fill speed setting
> > on the nozzle, usually the slowest speed.
>
> A variation of a half gallon or so on 300 miles isn't going to have a
> significant effect on the mileage.

FSVO "significant effect".

With a car that returns 30 mpg, your "a half gallon or so on 300 miles"
represents a difference of 5% or thereabouts. Heck, with some
econoboxes (of the European variety, at least) it could be as much as
10%.

"Robert Briggs" > wrote in message
...
> Jeff Strickland wrote:
> > Bruce L. Bergman wrote:
>
> > > Using the same gasoline pump means you will use the same fill
> > > nozzle, and are parked at the same level (or not level) attitude on
> > > the pavement. Oh, and you want to use the same fill speed setting
> > > on the nozzle, usually the slowest speed.
> >
> > A variation of a half gallon or so on 300 miles isn't going to have a
> > significant effect on the mileage.
>
> FSVO "significant effect".
>
> With a car that returns 30 mpg, your "a half gallon or so on 300 miles"
> represents a difference of 5% or thereabouts. Heck, with some
> econoboxes (of the European variety, at least) it could be as much as
> 10%.

If a car gets 25 mpg, and goes 300 miles, it will take a new fuel load of 12
gallons. If you cram in 12.5 gallons, the mileage will be 24 mpg, if 11.5
gallons then it will go to 26 mpg. This might be significant on a single
tank of gas, but over time it won't matter much.

The way to measure mpg is to fill the tank and reset the trip meter. Then,
on the next trip to the gas station, fill again and divide the distance
traveled by the quantity of the fuel load. Reset the trip meter, and repeat.

Unless one is doing a science project where the exact figures are critical,
then getting gas at Pump 1 on one trip, and Pump 6 on another won't make any
difference. Getting gas at one station (Brand X) and then getting gas at
another station (Brand Y) will have a greater impact than jumping from pump
to pump within the same station, or different stations of the same brand.

Robert Briggs wrote:
>
> Jeff Strickland wrote:
> > Bruce L. Bergman wrote:
>
> > > Using the same gasoline pump means you will use the same fill
> > > nozzle, and are parked at the same level (or not level) attitude on
> > > the pavement. Oh, and you want to use the same fill speed setting
> > > on the nozzle, usually the slowest speed.
> >
> > A variation of a half gallon or so on 300 miles isn't going to have a
> > significant effect on the mileage.
>
> FSVO "significant effect".
>
> With a car that returns 30 mpg, your "a half gallon or so on 300 miles"
> represents a difference of 5% or thereabouts. Heck, with some
> econoboxes (of the European variety, at least) it could be as much as
> 10%.

Has anybody determined what the original poster is going to do with this
information? Does he really NEED 3 decimal places accuracy?

--
Cheers, Bev
================================================== ==============
I didn't break it! It was doing that before I broke it... er...

The Real Bev > wrote:
: Has anybody determined what the original poster is going to do with this
: information? Does he really NEED 3 decimal places accuracy?

No, if the uncertainty was less than1mpg (that is, 28 means definitely
somewhere between 27.5 and 28.5), that would be fine.

The Real Bev wrote:
> Robert Briggs wrote:
>
>>Jeff Strickland wrote:
>>
>>>Bruce L. Bergman wrote:
>>
>>>> Using the same gasoline pump means you will use the same fill
>>>>nozzle, and are parked at the same level (or not level) attitude on
>>>>the pavement. Oh, and you want to use the same fill speed setting
>>>>on the nozzle, usually the slowest speed.
>>>
>>>A variation of a half gallon or so on 300 miles isn't going to have a
>>>significant effect on the mileage.
>>
>>FSVO "significant effect".
>>
>>With a car that returns 30 mpg, your "a half gallon or so on 300 miles"
>>represents a difference of 5% or thereabouts. Heck, with some
>>econoboxes (of the European variety, at least) it could be as much as
>>10%.
>
>
> Has anybody determined what the original poster is going to do with this
> information? Does he really NEED 3 decimal places accuracy?
>
Seems to me the way to eliminate the pump cutoff point problem is to
fill till you can see fuel in neck. Variation then should be very small.

Newbie wrote:
> The Real Bev > wrote:
> : Has anybody determined what the original poster is going to do with this
> : information? Does he really NEED 3 decimal places accuracy?
>
> No, if the uncertainty was less than1mpg (that is, 28 means definitely
> somewhere between 27.5 and 28.5), that would be fine.


A few years ago I did tests to see whether my Neon with spark knock
sensing would get better milage on premium vs regular. With ten fillups
on each, the variance on each set of measurements was 0.2 mpg. Regular
showed 0.1 better than premium, but since variance was 0.2 I had to
conclude no difference.

The fact that variance was so close on both sets of measurements (I
think it was 0.17 on one, 0.18 on other) indicates to me that at least
the way I was doing the testing was accurate (precise, actually) to
about 0.2 mpg.