Gas mileage saga continues (Sorry, QUITE LONG)

[Marteney, Steven J.]

Okay, verified in the archives a discussion of air fuel ratio.  Appears to
be the mass of the air divided by the mass of the fuel (not volume.)  If
this is INCORRECT, stop reading and let me know.   Now I also believe this
is the mass of INCOMING air divided by the mass of INCOMING fuel in order to
burn all the fuel and oxygen available, correct?  Yet again, if not, stop
reading and let me know.

So, that said, decided yesterday to do a little Diacom test.  Cruise set at
75mph.  Windows up, record some data.  Windows down, record some data.  This
really isn't the important stuff.  But here's what I did with the data.

I took 16 Diacom frames and averaged the reported pulsewidth (in ms), mass
air flow (in gps), and rpm.  For example, these values were 3.2ms, 38 gps,
and 2350 rpm, @ 75mph.  (I've spared you the actual numbers, like
3.327656... just to make it readable.  Originally done in an Excel
spreadsheet so round-off error isn't too prevalent.)  So, here's the math:

1) 22 lb/hr * 48psi / 43.5psi = 24.3 lb/hr effective injector flow rate
2) 1 / 2350 rpm * 60 sec = 25.5ms per revolution (I guess this would be
double-fire?)
3) 3.2ms / 25.5ms * 24.3 lb/hr * 8 = 24.4 lb/hr of fuel being used given the
injector duty cycle at that rpm
4) 24.4 lb/hr * (1000g / 2.2lb) * (1 / 3600sec) = 3.39 gps of fuel used
5) 38 gps of air / 3.39 gps of fuel = 11.2:1 AFR

Okay, so what did I do wrong?  Before I did this calculation I was
calculating gal/hr of gas used from the pulsewidth and mph of speed to calc
mpg.  What do you know, using that data I predicted 16.7 - 17.5mpg for the
windows down versus windows up condition.  So, I think the reported
pulsewidth from Diacom is at least somewhat accurate.  So, if the air usage
reported by Diacom is close, it says I'm operating at an 11-ish:1 AFR, way
too rich.  Now, if I need 30% more AFR to be at stoich, I should get 30%
more mileage (a big assumption) which means about 22mpg (baseline of 17mpg).
That's what guys with 350's keep telling me I should be getting but I never
seem to achieve.

So, am I way off base?  Is the oxygen sensor happily and incorrectly
reporting a good mixture even though it is way too rich?  I should note, the
exhaust typically smells rich.  Is the oxygen sensor really THAT bad?  It is
fairly new, 6 months old.  Before I go fooling the computer by changing a
commanded AFR to something like 17:1 so I actually get 14.7:1 in reality,
I'd like some opinions on this.

Signed,
Confused in Florida
(aka Steve)

Side note:  No more road tests for a couple of days.  AIR pump suddenly
seized this morning.  Now I know there are going to be replys that say the
pump was loading the engine.  Keep in mind, my BLM, TPS, and MAF reading at
cruise typically stays constant or has gone DOWN over a period of months, so
I don't think this was the case.  Guess it could be.
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org