OK, so key on engine off is where it should be, the the problem starts on ignition. Anyone have a clue what fault a fixed 40% and fixed 96% duty cycle indicates?

I have forgotten what car we are taking about?

89 M103

fixed 40 % says open circuit or short circuit to air flow sensor potentiometer or faulty. Possibly,fast idling speed. Test air flow sensor potentiometer. Test wiring, cis-e control unit.
96 usually points to an open circuit some where. Bear in mind this is my 560 manual.
m

woops next page 100 says: Lambda setting too lean. O2 sensor faulty short to ground. No voltage or ground at cis-e control or cis-e control unit faulty. lambda tester faulty.

things to do test overvoltage protection, test ground,test setting of lambda control and O2sensor signal cis-e faulty.

capn, I have found that the 96-97 reading was caused by my 4mm allen being over adjusted. and so my readings where all screwed up.
m

Mine only goes to 96% after it has been running for around 10 minutes. Then all of a sudden it will go back to 39.3%. Then it may go back to 96% and so on.

capn, thats what mine was doing. I also got this when i lost the eha current. But after adjusting the fuel mixture allen the problem went away and I get nice eha and duty cycle numbers. Still have the stall on acceleration but the electronics work.

m

The 100% reading (96) indicates the car is too lean. The fixed reading is either a result of that or the airflow pot as described.

While the reading is 40% there isn't much you can adjust, but when it is 100% try pushing down on the airflow meter flap. This should richen the mixture and if the O2 sensor is functioning the duty cycle should drop before you push so hard it starts running bad.

You should also disconnect the O2 sensor while reading 100%, it should immediately drop to 50%. I think the O2 sensor code 50% is the prevaling code. So it might even go to 50% when it is 40% if you disconnect.

Your explanation of grounding the O2 through your body is wrong! The test goes like this: simulate lean by grounding the sensor the duty cycle should head high. Simulate a rich mixture, run battery voltage through your body to the computor lead from the sensor: the duty cycle should go low. Running the O2 sensor input to ground through your body probably won't do anything.

The EHA current should go to +12ma when shorting the O2 input to ground and the current should go -12ma when the term sees battery voltage through ones body. (DO NOT HOOK battery voltage straight to the sensor input - we are trying to simulate 1-2v)

I have changed the sign convention twice now after rereading this. I think I got it right.

I would suspect either you are way off on mixture, or the O2 sensor is bad. One way or another, you aren't in closed loop mode (see steve's other posts, too!).

Don't adjust much at a time on the fuel distributor, and wait a while between adjustments -- I had the TE in almost the same state because I was expecting it to react too quickly. Takes 10 sec or so for the O2 sensor to register mixture changes at idle! Also, the pressure on the adjuster screws things up, I had to release the spring for it to stablize.

Mine is now at 50% duty cycle with constant variation, although I still have some intermittant miss vibration. Someday will take it down and stick the CO meter in the tailpipe and see if the mixture is out of synch with the controller. In the mean time, I'm running some injector cleaner in it and thinking of injector seals.

Peter