I recently (finally) completed some air fuel tuning on my conversion
(http://my.tbaytel.net/guskers) which uses digifant II injection.  Given
cold weather, short hop mileage (open loop operation) which was poor, and
closed loop operation mileage at 27mpg (auto, westfalia), I was fairly sure
that the AFM's base setting was rich.  With the o2 sensor unplugged, I was
getting .7 to .8V from the o2 sensor itself at hot idle.  The CO adjustment
screw would not bring this down more than .1V.  The exhaust smelled pretty
rich on cold idle.

I popped the AFM cap and adjusted the cogged wheel 2 teeth clockwise to get
the O2 sensor reading around .5 volts.  The readings from the o2 sensor
(disconnected from ECU) now became pretty sensitive to the CO adjustment
screw, fluctuating appreciably with small CO adjustments.  I also re-
adjusted the base idle to 1000rpm with coolant sensor disconnected...it
returned to 850rpm or so with the connector hooked up as per spec.

During driving, my dash mounted lascar meter
(http://my.tbaytel.net/guskers/gauges.html) shows .5 volts (as expected)
when cold, and then shows rapid fluctuations between 0 and .8 volts once
warm.  At 25 to 30C it goes to closed loop in a minute or two. Above
4000rpm at full load it reads a solid .65 to .8V.  Previous to the
adjustment it seemed to "peg" at .8 Volts.  I haven't tried releasing
propane into the intake to find out what the O2 sensor's max reading is.  I
haven't tested performance yet with my G-tech, but high rpm highway
performance seemed noticeably brisker.

Can anyone critique my technique here?  I'd hate to blow up my motor.