Your CO is very high. You have a mixture problem!

The first thing to check is your O2 sensor. There have been numerous posts on this forum about checking the O2 sensor output wave form and "duty cycle" of the EHA. Don't know how good a home mechanic you are, but with a digital multimeter with a scope or duty cycle function, it's not that hard a job. The other alternative is to just replace the O2 sensor apriori (it's the most likely culprit) or find a COMPETENT mechanic who understands the KE system to do the diagnositics and repair. The fact that your mechanic thinks 13.9 percent CO2 is okay gives him zero credibility in my book.

Notice that the sum of your CO2 plus 02 is 15.0 percent. The emission control system on your engine operates by maintaining very close to stoiciometric mixture under all but WOT conditions. Native emissions out of the engine should be low, especially CO. The three way converters then reduce NOx creating free oxygen, which, along with the very small quantity of oxygen left over after combustion is available to oxidized CO and HC. With a stoichiometric fuel-air ratio there is only a fraction of a percent CO so most of the available oxygen is available to oxidize HC.

The following numbers are from the latest test on my '88 190E 2.6 last February - same basic emission control architecture as your 300E - no air pump, no EGR, just stoichiometric air-fuel ratio control, igntion timing control, and three way converters (oxidize HC and CO and reduce NOX).

15 MPH allowable (limit)

HC 113 (116)
CO 0.36% (0.74%)
NOx 261 (791)
CO2 15.3
O2 0.1

25 MPH

HC 58 (91)
CO 0.18 (0.62)
NOx 188 (730)
CO2 15.3
O2 0.0

On the suface it appears that my mixture is a bit rich, but my O2 sensor output and duty cycle are within acceptable limits. O2 sensors can drift and become less accurate over time, especially if they are not very hot. Same applies to converters. I still have my original O2 sensor (70K miles) and may have to replace it eventhough it's behavior appears to be okay.

Note that the 25 MPG emissions are better and all the O2 is consumed. This is because the converter heats up on the 15 MPH test (which has a higher relative load than the 25 MPH test) and is more efficient during the 25 MPH test.

One trick I know of is to have the car tested on a rainy day and specifically ask the tech to do the "tire drying procedure". The test procedure allows this additional step to dry off the tires prior to the loaded test when emission sampling is done. The engine is run in gear at no load for about 1 minute at about 25 MPH, but this is enough to heat up the converters and O2 sensor for higher operating efficiency.

I did this two years ago and 15 MPH HC was 87. After a thorough diagnosis I thought I would be okay, but I just squeaked by.

Next time I will go on a rainy day and ask them to run the tire drying test.

BTW, our 103 engine family has been identified by the CARB as a "high emitter family", and I have noticed that many onwers are having problems with marginal failures, even the later versions that have air pumps. Unfortunately, our emissions control system does not have as much margin for degradation as many other engine families.

The actual limits have nothing to do with the absolute standards that our cars were certified to. The field test limits are based on statistical sampling and analysis, and this is were I have argued with the CARB, especially with marginal failures.

Once you get your mixture problem fixed you will still have to watch HC.

It's basically a game we have to learn how to play! I'd like to hear a follow-up on the repair and final test numbers as I'm still researching the high emissions issues on the 103 engine family.

BTW, how many miles does your car have and do you know how many miles on on the O2 sensor?

Duke

P.S. I don't think the A/C is supposed to be on or maybe the test procedure does not specify - I'm not sure. Having it on prior to the test places a little load on the engine, which should help keep the converter and O2 sensor hot, and it will mean a little more load on the engine during the test, but it should not materially affect the emissions.