O2 sensor and 1223 code revisited

["KARL MAXON" <KLMAXON@worldnet.att.net>]

To all:
I've had a few inquiries regarding the 1223 code I had on my '91 164L and
what the solution was.  To recap:

My original message:
On my '91 164L, I've gotten both a 1223 code (lambda probe inner adjuster
reaches min or max value of correction), and a 1224 code-once (lambda probe
transmits to control unit a tension > 1.009v or <0.0928v).

I have reviewed all the available CD ROM manual/technical bulletin
information (Motronic manual troubleshooting guide, TB 90.01, TB 90.03).  I
did find that the lambda probe "heater" connector was full of water and
corroded!  All other items in the "tests" checked out OK, so I'm highly
suspicious of the O2 sensor.  Would your collective experience out there
tell you the O2 has failed?  I cleaned the connector, used some Stabilant,
also some dielectric grease for packing, cleared the codes, yet upon
driving the error codes returned.

The car appears to be in fail-safe mode at the moment.  In TB 90.01 it is
noted that if the heater connector gets water or condensation in it that
the connection can short out, thus "fouling" the sensor with carbon.  All
they indicate in the manuals and TB's is "replace the sensor".  No
indication is given on what to do if a sensor is "fouled" (except replace
it).

My question is:  Are O2 sensors that are "fouled" cleanable (like a fouled
spark plug)?  If so, this could save $$$ for new sensor, and time to
replace it.  If cleanable, are there recommended procedures?  The current
probe has 45,000 miles on it.

My car exhibited stumble start up and running after it would sit for a few
hours (after being driven).  The symptoms came on gradually over a 3-4 day
period, until now what seems to be fail-safe mode and stumble start up and
poor running under most conditions.  As indicated above, I checked all the
air boots, hoses, and wiring several times, and the car has newer cap,
rotor, plugs, etc which I also checked several times.  It just seems that
both error codes combined point to the O2 sensor, and that it may be
fouled.
End

The final solution was to replace the O2 sensor, as diagnosed at the shop
as follows:
The repair shop verified that the O2 was working, and COULD induce the O2
sensor to respond by pulling the vacuum hose to the cruise control off
(easiest to reach, and wouldn't effect fuel pressure).  This threw them off
for awhile, since the readings
from the tailpipe were off the chart.  The clue was when they REMOVED the
connection to the O2 sensor, the computer would not go to a "base" setting,
and would stay too rich (the car was running super rich with the O2 sensor
hooked up too), and this led them to investigate a lot of other systems for
potential problems.

The cat conv. was "flooded" as there was lots of fuel and rotten egg smell
(the car had been running rich).  ALL other systems checked out perfectly,
and they spent a lot of time investigating every possible item (throttle
body sensor out of adjustment, throttle cable, vacuum leaks, fuel
over-pressure, fuel injectors, all ignition components.

Their solution:  Since the O2 would be essentially at one position/reading,
and would only move when induced by a vacuum leak (only at a barely
perceptible level), they replaced the O2 sensor.  Upon firing up, it took a
few minutes for the cat to heat up and clean out, but from there on all
readings from
the tailpipe and probe were BETTER than they have ever been on the car
since I've owned it.  The car would correctly go to base when the O2 was
disconnected (albeit a little "high" at around .7), and when connected
would fluctuate around a mean of .5 

We looked at a print-out from an earlier tune-up (at 95,000 miles), and
while within
acceptable range, in hindsight the readings were not that great (HC=152
CO=.53 CO2=14.64 O2=1.15 Lambda=1.03 AFR=14.9), and we
should have replaced the O2 sensor at that time.  Note that the sensor had
only 45,000 miles on it at failure.  So, as indicated by the tech bulletin
(for the 1223 code), if everything else checks out, replace the O2 sensor -
which turned out to be correct!!!
Readings upon replacing the O2 sensor: HC=9 CO=0 CO2=15.28 O2=.63
Lambda=1.02 AFR=14.9.

I'm suspect of myself for the failure to some extent as I had done a valve
adjust and belt replacement 2 weeks earlier, and had cleaned and used
Stabilant on every single connector I could find, as well as cleaning the
Idle adjust actuator, air flow meter, throttle body, intake tubes etc on a
car with 105,000 miles.  Could it be that all other  items were optimized
to some extent with the better connections, which placed the responses from
the O2 sensor (that may have been marginal) somewhat out of range?  

Some respondents indicated the O2 sensor could be cleaned, other suggested
not.  Either way, if the O2 sensor is pulled, if an inexpensive one is
available it may not hurt to replace it.  Note that Fred Di Matteo cautions
that a thorough diagnosis should be made for any problem before parts are
substituted!  This is why I ended up taking the car to a shop that could
thoroughly diagnose every system better than I could with the tools I had
available.

The Sports Car Shop in Eugene did an outstanding job (and lots of extra
attention) to resolving the problem, and thank you to all who responded
with your messages and advice!  While a little long, I hope the above will
be of use to others who encounter the same issue.

Karl Maxon
1991 164L
1972 Spider