On Sun, 17 May 1998 20:25:35 -0500 Steven Broomhead said:
>If my tanks holds 10 gallons, say I let it idle in my driveway for 3 hours
>and then start the trip.  My tank would run dry at 225 miles.  My true mpg
>would be 22.5, BUT I used x (unknown qty of fuel) during the 3 hour idle.
>Since the engine does not idle on pure air, the amount of fuel used is
>negligable.

in theory. but in test on my 86 van, with a trip computer installed (that
measured the amount of fuel used, and was pretty darned accurate ... down
to 0.1 gallon), i found that that particular van used 0.6 gallons per hour
idling in the driveway. yeah, i know ... who'd REALLY be dumb enough to
do this. well ...  it SEEMED like a good idea at the time. :) but i was
fooling around with the trip computer and the new bus, my first water-cooled,
so i wanted to see what it would get. what i found was this:
Bentley says: 18 mph per 1000 rpm.
1986 data:    16.4 mph per 1000 rpm. tires, friction losses, and air drag,
                   i guess.

so i had a table like this:
         calc                 range in
rpm      mph     gph    mpg   miles*
3000     49.1     -      -      -
3200     52.3    2.3    23.1  346
3400     55.6    2.4    23.6  354
3600     58.9    2.7    21.2  318
3800     62.1    3.2    19.2  288
4000     65.4    3.3    19.8  297
4200     68.7    3.7    18.8  282
4400**   71.9    4.3    16.8  252

-  = no data. people were trying to run me off the road at that speed. i also
     couldn't get up the very small hills on my "test track" of a pretty flat
     area on the interstate highway just west of here.
*  = assuming you ran 15 of the 16 gallons in your tank at this speed.
** = only got to run one short test at this speed, so i didn't have enough
     data to average (like the other speeds: they were all averages of three
     test runs). remember, this was back when 55 mph was the interstate
     speed limit. and a bus makes a very good radar target!! :)

now, in real-world driving, you can't just sit there on cruise control all
the time. people slow down in front of you, you can't pull out when you need
to get around them, you have to stop for red lights and all sorts of things.
dirty bus = higher drag. tires get low on air = higher drag. that sort of
stuff. so what i found was ...
- best speed for best mpg:   3600-3800 rpm. below 3600 got better mpg, but
                                       didn't have enough omph to get up
                                       most hills. and hills REALLY dragged
                                       the mpg down (the trip computer would
                                       also show instantaneous mpg).
- worst speed for mpg:       above 4000 rpm. the higher rpm you drove, the
                                       worse the mpg got.
- best ever mpg:             24.3 mpg.
- worst ever mpg:            14.9 mpg.

>In short, no matter what my driving habits over the course of a year my mpg
>averaged 25 MPG +/- 1 MPG.

then i would suspect that
- you drive slower than most folks on the road.
- you drive mostly highway miles
- you keep your tires pumped up
- you keep your bus clean and shiney.

... or you're not keeping the records right. on mine, i had to make sure that
i filled up the tank to the same spot in the filler neck. otherwise, there
could be as much as a half gallon or more difference in the amount used to
fill up. plus the time of day makes a difference ... fill up at night and
you get a little colder gasoline, which is more dense, so you can cram a
little bit more into your tank. fill up at 2.pm in the afternoon, and the
gas is pretty warm and expands to fill the tank with less gallons. now,
with the small 16-gallon tanks we have, it's not gonna be a whole lot...
maybe not even noticeable, like a few ounces. but all these factors can work
against you ... and if you depend on the pump nozzle kick-off (when it stops
pumping), you can also vary quite a bit ... not all pumps are the same.

my point being that even if you do it super-critcally, you can STILL come up
with differences. the reason i did three runs at each rpm and averaged them
was because at different times of the day, in different weather, i'd get
different results. oh, they were close, but not exact.

joel