Another CO Adjustment Question. - PeachParts Mercedes-Benz Forum

300EVIL · #1 10-09-2003, 10:44 PM

After correcting my CO adjustment a month ago everything was working just fine until a week ago when I noticed the engine was hard to start, hot or cold, and a slight rough idle as well. I decided to check my duty cycle reading again this evening and found it to be sitting at 92% flat!!

Fortunately it is not running that lean, I can tell by the smell of the exaust. Is this a trouble code? I have tested the battery voltage, the OVP relay, the EHA and the idle air valve,,,, All of these seem to be functioning properly. What else should I be looking at??? Your suggestions will be greatly appreciated.
Thanks!
Adam

300EVIL · #2 10-10-2003, 08:54 PM

TTT

stevebfl · #3 10-11-2003, 11:16 AM

Usually my first step is to push down on the airflow plate with my finger and see if the system responds to the richer mixture.

I then disconnect the O2 sensor and verify the ability to give a 50% fault code. If it does then I would try readjusting to the proper point.

One must understand that there is no significant difference inmixture between closed loop at 90% and closed loop at 50%. If the system reached the end of compensation somewhere in the high 90 percentile, immediately it would only be slightly different mixture than at closed loop 90%.

Mixture is held to a very tight tolerance when in closed loop and the car can run perfectly significantly above or below what is achieved in closed loop.

richard28 · #4 10-14-2003, 10:55 AM

Steve,
Does the 90% reading mean too rich or too lean? I also have a fixed 90%, but when I depress the air flow plate the reading increases for the one second before the car died. Since the reading started at 90%, wouldn't that mean the car was already running rich? Thanks.
Richard

stevebfl · #5 10-14-2003, 11:17 AM

Quantitatively, 90% is the response to a lean mixture. it is a rich response but the reading confirms that correction was not achieved. Thus you should still be lean or you would no longer get the response.

300EVIL · #6 10-16-2003, 09:44 PM

Okay,
I umplugged the o2 sensor and I get a 50% reading I also unplugged the EHA and it returned a 30% reading. I checked for vacuum leaks as well and found none. the temp sensors are within spec as well. Any other suggestions will be greatly appreciated.
Thanks,
Adam

stevebfl · #7 10-17-2003, 08:19 AM

So you have a system that responds! The system is saying: I see lean go rich. It is never getting there OR the sensor is lying.

I would gently apply pressure to the airflap and increase mixture. If the car instantly runs poorer as the flap is pushed you are obviously not lean. I would begin to distrust the sensor.

Generally speaking if you aren't stuck on a duty-cycle fault code, a high reading will mean a lean mixture and adjustment is what is necessary. If the thing doesn't respond then I would suspect a grounded O2 sensor.

300EVIL · #8 10-17-2003, 08:06 PM

Hi Steve,
Thanks for the info!!!
Yes, I already tapped on the air flow plate and the engine either cuts out or runs poor with a millimeter or more of actuation. I tested the O2 sensor today. the heater voltage and filliment seem to be present however the O2 voltage output is only a few millivolts I am assuming the sensor is toast considering it is supposed to put out between 0.5 and 1.0 at normal operating temps. Would This analysis be correct??? If so, Why are my not getting a 50% duty cycle reading if it is bad? Could the abrupt failing of the O2 sensor been caused by running at a rich mixture for two years and then setting it back to normal operation? BTW, I have a hard cold start,, about 2-3 seconds on the key before ignition,, would a bad O2 sensor cause this condition???

Thanks a lot for everyones "free" help here! It is appreciated vemuch. I would pay a subscription fee if there was one. I do try to donate the knowlege that I have gained to other members though.

Thanks Again!
Adam

stevebfl · #9 10-18-2003, 09:15 AM

The sensor can provide up to 1v output starting at zero volts not .5 volts. As a result a few millivolts is an acceptable reading and the system won't fault it.

Yes, a sensor becomes coated and inoperative if left in a over-rich mixture for extended time. The voltage output should be checked with the engine running and after holding the engine speed to 2000rpms for at least 30 seconds (my own criteria), as it needs top be hot. Without a good heater turned on the O2 sensor won't get hot enough at idle. If you can burn through the crude and get the first reading the sensor will probably clean itself if closed loop can be established.

To test the sensor disconnect from harness and measure the output voltage as one slightly changesthe fuel mixture. If the voltage is low then just a small push on the airflap should bring an instant change to near 1v. Remember that the O2 sensor only reads in the very lean range of mixture probably from 14/1 to 15/1 air/fuel ratios. A really rich mixture only reads the same 1v as does the 14/1 barely rich mixture. In %CO the sensor reads mixtures from 0 to 1% trying to keep the mix about .5% CO.

I diagnosed and repaired a Ferarri with straight K-jet for another shop yesterday and we left the mixture a little lean at around 2.5% CO. It was a seventies model and had no emissions on it. It probably wouldn't run at .5% CO.