Date:    Fri, 14 Mar 2003 13:54:05 -0500
From:    David Brodbeck <gull@GULL.US>
Subject:

On Fri, 14 Mar 2003, Mark Keller wrote:

> The Digifant Fuel system works largley by monitoring  input from the O2
> sensor.  By monitoring the O2 sensor with an averaging multimeter you
> can see the average fuel mixture theengine is actually getting. Because
> the engine is constantly getting a little less and a little more than
> this average, it probably isn't even producing the same power it could
> if it got a constant mixture.

What you're looking for seems almost impossible in a vehicle
without a throttle position sensor. Unfortunately, the AFM reading
is dependent on *both* the throttle position, and the RPM, and thus
is insensitive to engine load. Additionally, since there is no
manifold pressure/vacuum sensor, the ECU is left with *no* way of
knowing what the throttle position is.

*IF* the throttle position is known, the easiest route to what
you're looking for would be to program the chip with the optimal
fuel/air map algorithm (i.e. the typical 17:1 mixture), adjusted
based on the O2 sensor output. Additionally, that algorithm could
incorporate an "enrichment factor" calculated from comparison of
the engine RPM versus throttle position, integrated over a set
period (similar to a typical PID algorithm). A slow rate of RPM
change, relative to the throttle position, would indicate heavy
load and result in additional enrichment. A quick rate of change
would indicate no/light load, and the "enrichment factor" would be
1 (i.e. the non-adjusted basic fuel/air map). Using this scenario,
you could achieve maximum power under heavy load, with minimal
impact under light loading conditions, but overall the fuel
efficiency and emissions would still suffer.

Keith Hughes
'86 Westy TiiCo, Marvin