02 Sensor Mixture Testing

[echrisconnor]

After my recent rebuild, I've been having a hard time getting my mixture in the ballpark and am trying to figure out a simple system of testing the initial mixture. My intent is to get it close, then borrow a Gunson for a final test.

This is for an early CIS car with no 02 sensor, but I used to have a sensor in my old headers. My new ones don't have a bung in place, but I still have the sensor and an 02 mixture gauge.

Here's my question. Can I temporarily put an extension on one of the outlets from my sport exhaust with the 02 sensor and get a basic reading? The headers and sport exhaust are a farily direct shot out of the exhaust with plenty of heat. I realize the sensor should be close to the exhaust port, but might this still get a vaguely accurate reading?

Here's a picture of my proposed plan.


I'd join a short length of exhaust pipe to one of the exhaust pipes to make this extension. It's just for testing purposes, and not a permanent attachment.

[targa80]

If you have the CIS with Lambda control box and O2 sensor then disconnecting the sensor puts the system in open loop mode. In open loop mode the Lambda unit puts the frequency valve in a 50% duty cycle. Temporarily inserting the O2 sensor will allow you to optimize the mixture and the frequency valve will be adjusted between 45 ~ 55 % duty cycle depending on the load conditions. To remove the O2 sensor will just put you right back into the 50% duty cycle. Why not just optimize the mixture in the open loop mode since that is the way you will be running the car.

[echrisconnor]

Pat, no this is a non-lambda car for which I used to have an 02 sensor mounted and hooked up just to a gauge just so I could monitor my mixture. It's not connected to the CIS system in any way.

My new headers don't have an O2 bung and I just want to get a snapshot of my mixutre via the 02 sensor and gauge I have. My question is whether temporarily moun ting the 02 sensor to an extension of one of my muffler outlets will give me a vaguely accurate reading. This will be moitored on an 02 guage and will not be interpreted by the CIS system.

[fireant911]

echrisconner,
I have been thinking about your proposal over the last couple of hours and reviewed my collection of CIS books in my ever-growing Porsche library once I arrived home. My conclusion is that you will not be able to obtain 'vaguely' accurate readings from this setup.

The pre-Lambda CIS systems have a higher target CO level thus you will first have to determine the target voltage reading that would be acceptable for your CIS K-basic system. The location of the O2 sensor, as I understand it, is pretty critical in obtaining accurate readings - so the temporary scenario you are considering may not allow good readings. I have attempted to set the basic mixture on my car with the O2 sensor disconnected (loop open mode, as in your car) while reviewing the readings from my Westach A/F monitor and have not had the greatest sucess with this method. My recommendation is that in just a couple of minutes, with the proper gas analyzer, will allow you to properly and accurately set the proper mixture.

[echrisconnor]

Thanks Daryl. I have a couple questions. The CO which is monitored by a gastester is the key measurement, but the 02 measured by the 02 sensor should still be close to 14:7 right?

Also, from what I know proper placement of the sensor is so that it can get good and hot, for accurate readings. In looking at the placement of the sensor in my BMW, it's down low, following a straight 6 just before the CAT. This is probably a longer distance than through my headers and out the exhaust. Also, this is a gutted sport exhaust like in this picture - it's a pretty straight shot through.

I agree, it's a quick easy process to go and borrow one, but I'd rather get this sorted to some degree and then use my borrowing favors for fine tuning.

[fireant911]

echrisconner,
I do not think that your target A/F ratio is 14.7:1 and my reason is simply that your CO value is between 1.5% to 3.5% whereas my CO target value is 0.4% to 0.8%. Whenever the mixture is richened the CO content increases so I do not believe that 14.7:1 (which converts to approximately 0.45 V from the O2 sensor) is proper for your car.
There are other methods that I have seen mentioned on this board for adjusting the basic mixture that may get in in the ball park. One is the method in which you remove the air filter cover and air filter. Then raise and lower the sensor plate ever so slightly and that the idle should drop if the mixture is correct. Please see post by kstylianos in the link below:

AFR meter or Gunson ??

[RarlyL8]

You don't need to use an A/F meter for a ballpark adjustment.
Simply warm up the engine and, using small incremental adjustments, turn the mixture screw rich until the idle begins to hunt. Back it off a smidge. You are now close enough to borrow the gas tester.
The best thing to do if you can is just wait until you can use the gas tester. Other than that the obove referenced post will do.

[echrisconnor]

Thanks all. I've already tried the throttle plate method and found what I believed to be the "happy place." I was thinking that the 02 sensor might get me an even better reading -- but possibly not. Since emissions isn't an issue, maybe I'll just stick with the throttle plate system until I go borrow the Gunson.

[Paulporsche]

Echriss,

If you do use a gastester later on would you let us know what your final "happy place" setting is (CO%), and if it is any different from the setting you got from using the plate raising method.