Another O2 sensor question...
 

In my ongoing efforts to track down the source of my rather sudden rich mixture and high oil consumption, the last thing I could think of was replacing the O2 sensor. My original one was 85k miles old and I believe I used silicone on its threads when I installed my test pipe. So having read here about how silicone can kill the sensor, I figured I probably got some on my old O2 sensor. My rich mixture started as soon as I installed the test pipe. I had a vacuum leak fixed and then the mixture reset. She still smelled very rich and used about 1 qt. of oil per 500-600 miles. Before all this she was using half as much oil. I mean it doubled almost overnight.

Anyway, to the point... 10 days ago I replaced the O2 sensor. I had topped off the oil (yes, I know how to check and add oil in a 911) about 25 miles before the O2 install. Afterwards, it still smelled rich. But yesterday, with about 415 miles on the new O2 sensor, I was only down 1/2 qt. of oil. So does it take a while for the new O2 sensor to start working, or does it kick in right away? I've driven only highway miles since then - from DC to NJ/Philly and back. It looks like my oil consumption has gone first from 1 qt. per 1000-1200 miles, to 1 qt. per 500-600 miles and now to 1 qt. per 800-900 miles. I won't rest until I've gone down another 1/2 qt. and measured it and the mileage. But I'm wondering if I may have somewhat fixed my problem here. How to tell? Sorry for the long post.

An O2 sensor should work as soon as it is heated up. IIRC there is a heating element within this O2 sensor. Hence, it should work within 1 MAYBE 2 minutes of running the car.

BTW: How did you check/adjust the mixture???
Did you (or someone) drill the MAF (barndoor) sensor to get to the mixture screw?

Also ANY silicone consumed within the engine, or in the exhaust that is NOT O2 sensor safe can (and probably will) destroy an O2 sensor.

A new O2 sensor starts working as soon as it warms to operating temperature.

I hope you cleaned the threads of the test pipe thoroughly with solvent before installing the new one. Replacement Bosch sensors come with copper-paste anti-sieze compound on their threads.

I did not clean the test pipe threads, but I did not see anything on them and I doubt anything there came into contact with anypart of the sensor other than its threads, on which I did use the copper lube this time. I had a pro-wrench adjust the mixture.

Silicone has an insidious way of getting on things that are far...far away from the source. Threads should have been free of silicone when the new replacement sensor was put in.

--Wil Ferch

If your old O2 sensor got contaminated and did not create a proper output, a U.S. 3.2 will run very rich - too rich. There was a technical note by Bruce Anderson in the recent issue of Excellence where Porsche stated that indications displaying varying degrees of oil consumption in a 911 were due to the amount of gasoline mixed in with the motor oil. Short trip driven motors, and those that ran rich had high levels of moisture and gasoline in the motor oil, which did not burn off until a driver had taken his car off for a long drive which brought up the oil temperature enough to vaporize the contaminants. The driver would then notice what seemed like a sudden increase in oil consumption, all the meanwhile, oil was just being consumed at a constant rate.

The other thing may be that since you replaced your cat with a test pipe, you may have smelled a rich exhaust, because the cat was no longer in the chain to catalyze the exhaust. It may have been running a little rich before with the cat. Properly tuned and running though, you should not be able to smell any rich running conditions, even without a cat.

Richard,

I'm far from a educated source on this matter, but I would run another 1-2K miles before trying to figure out the data. My 86 with 110K was consistently burning 1 qt/600 miles and had some smoke for the 5 months I owned it. Well, I had the oil changed before a road trip and the shop discovered that my O2 sensor connnection was bad. They reconnected and here's what I've seen over the last 1.5K miles:

- Improved average MPG from 20 to 23-24.
- Rough idle and slight surge between 3-4K RPM
- No more smoke...well, not nearly as much that I can tell

and are you ready for this..

- 1 qt/1100 miles.

This in spite of another oil return tube going AWOL on me. I was going to replace my O2 sensor but after reading some recent posts it might just be acting normally. I don't know if any of this is related to my extended freeway miles over the road trip, change in weather, or just "that time of the month". In any case I haven't dropped my engine like you so what the hell am I talking bout!

Good luck and keep us posted. I for one am very interested.

All this leads me to another silly question. How can I test my O2 sensor? Is there an in-line test I can rig up, and if so, does it have to be done with the car hot? I bought the cheap, aftermarket 3-wire O2 sensor on the recommendation of a very knowledgeable Pelicanhead who also has one. It could be that I messed up the soldering/crimping when I had to connect it to my existing wires. But how do I determine for sure that it's working?

Richard, my car also smells slightly rich since fitting the test pipe. I put this down to the cat not being their to burn off or "filter" all the crap. Maybe I've got a dodgy sensor too?

Did the Euro spec Carrera's run an O2 sensor as they were not cat equipped? If not the DME must have been different, mustn't it?? If that's the case could I get shut of my sensor and modify whatever is different on the Euro models? (Not sure what I'm on about here!!)

Nathan

Nathan:
Some clarification is in order:
- USA ( and some...select..other countries, Japan, for one) use a DME for 3.2's that uses a cat and an O2 sensor.
- Euro cars...in *general* do not use a cat ( have a "pre-muffler" instead)..and no O2 sensor. Some Euro cars, optional for German market, have an O2 sensor and cat, run lead-free gas, and get a "tax" break. All in all...very much like a USA or Japan car, but ( I believe) with the higher compression ratio of the standard Euro car.
Each of these markets has a unique DME part number and/or certain wire functions of the DME activated/deactivated ( jumpered).


-- Wil Ferch

Wil, my (Swiss) car's engine is a type 930/25, which I believe is exactly the same as US spec - 217hp. Does anyone know if any DME functions can be altered to allow removal of the O2 sensor? (sensor seems to cause more trouble than it's worth to me!)

Nathan:
My suggestion?... keep the O2 sensor...just make sure all is up to snuff, including replacing the sensor when necessary. When I went to a Euro-premuffler for my USA car...I purposely added an Os sensor bung to keep the closed-control loop active. Actually works vert well, plus remember...at Wide Open Throttle, the O2 sensor circuit is deactivated anyway..and a pre-programmed power ( fuel rich) signal is used.

--Wil Ferch

You can test your sensor.

You mentioned it is three wire, so that means it is heated. Two wires for the heating element and one for the sensor. You need to find out which one is the sensor wire.

Remove the connector on the O2 sensor and connect a wire to the sensor wire (not the heater wires) and run this to the cabin. Connect the red lead of a voltmeter with a high input impedance (usually all digital meters are ok) to this wire. Connect the black lead of the voltmeter to the car ground somewhere.

Now go for a drive. Look at the voltmeter if it is reading less that 0.5volts this means you are running lean. If it is reading more the 0.5v then you are running rich. It won't read much more than 1volt.

Ideally it should toggle above and below 0.5v indicating a perfect mixture.

For more info, read this Pelican tech article:

http://www.pelicanparts.com/techarticles/mult_air_fuel_monitor/mult_air_fuel_monitor.htm