K-Jectronic adjustment

["Bob Rossato" <bob.rossato@att.net>]

I need a little help from the fuel injection experts on this list.  Subject
car is an '86 Scirocco 8V with the K-Lambda CIS.  I recently went for
emissions testing and failed the NOX portion of the test.  HC and CO were
well within the limits, however, the NOX emissions were ~2800 ppm compared
to an allowable 1400 ppm.  OK I know that high NOX is due to the car running
lean but the problem I'm having is in adjusting the Duty Cycle on the
Frequency valve.

I jumpered the fuel pump with the engine off to go through the regular
diagnostics and all the default settings seemed fine.  Cold Enrichment -
80%; Full Throttle enrichment - 65%; O2 sensor disconnected - 50%.  When I
started the car and let it warm up the Duty Cycle would vary from about 25%
to 40%.  Now, on my duty cycle meter 100% corresponds to full lean and 0% is
full rich so I assumed the meter was reading backwards as the enrichment
settings from Cold start and full throttle are 80% and 65% (which agree with
the Bentley values) , and during normal running duty cycle measures 25-40%.
So if the car is truly running lean the readings make sense.

When I tried to adjust the mixture however, it got a little weird.
Clockwise rotation of the adjustment screw is supposed to enrich the
mixture.  However, when I tried that the duty cycle went further towards
lean (rich on my meter). The Bentley confirms that clockwise rotation
enriches the mixture and by inspection that makes sense as clockwise
rotation will raise the plunger in the fuel distributor relative to the air
plate.

So how is it that when I activate the full throttle enrichment switch the
duty cycle goes to 65% and when I enrich the mixture via the adjustment
screw the duty cycle goes in the opposite direction towards 0%?  What am I
missing?  Do I trust the meter and go counterclockwise though it would seem
that is actually leaning the mixture  out more?

Also the Lambda system seemed to act erratically in that on occasions it
would just slowly move over to 100% duty cycle and stay there. The frequency
valve would obviously stop cycling at this point.  If I disconnected the O2
sensor it would revert to the 50% default.

I have changed the thermo switch for the cold enrichment (resistance values
were in the M-ohm range when cold compared to spec of 0.5 ohm max.) and put
in a new O2 sensor but the readings are the same, though it hasn't gone
towards full rich yet.  I know the thermos switch isn't the culprit as that
just affects cold running but it needed changing anyway.  Any other sensors
that will cause the ECU to affect mixture?

Any help would be appreciated.  I need to bring it before Nov 3rd for my
free retest.  I may just disconnect the O2 sensor and hope that 50% duty
cycle gets me through NOX.

TIA,
Bob R.