CIS mixture with Oxygen Sensor, questions
 

Hi folks,

I have my car, an 82 SC, in a pretty good spot now. Just rebuilt the transmission, new catalytic converter, new oxygen sensor, valve adjust, etc... I just set the timing/idle/mixture, and it now starts well and runs well. I am very pleased!

But I have a few questions that hopefully some of you could help me with. To set the mixture (which didn't really need much of a change at all) I used two tests, neither with a proper gas analyzer:

1) I pushed up on the air-flow plate, the RPMs surged then died as I continued pushing. As I understand it, this tells me that my mixture is correct or close to correct. My engine is telling me the same thing.

2) Then I connected a voltmeter to the oxygen sensor, and used the chart on this thread (http://forums.pelicanparts.com/porsche-911-technical-forum/179654-cis-basic-mixture-adjustment.html) to infer my mixture. At idle it was reading ~ 0.1V, which is extremely lean! But when I rev it it goes to 0.85V which is rich, but about where it should be according to that chart.


How would you guys interpret the 0.1V at idle?

Thanks,

Babak

PS I should add that during this test the oxygen sensor was not hooked up to the rest of the system. It now is, for the first time since I've owned this car, and all is running well.

Best to adjust with a pressure gauge and analyzer. The lean condition at idle isn't the best situation. Have you looked for vacuum leaks?

As an '82 you should have a lambda box and frequency valve. If so, there is a grn/wht test lead to connect to a dwell meter. Then it's just a matter of setting the mixture to get the freq valve dwell to 50%.

I'd also make sure you have a steady 15" vacuum at idle. And before freaking out looking at the various vacuum can doohickies in tracking down possible leaks, it may be as simple as the frayed ends of the vac lines at the distributor and throttle body.

I'll look for the test lead, so I can easily make the measurements with the Oxygen sensor connected. Thanks! Do you know what voltage corresponds to a dwell of 50%?

No visual signs of vacuum leaks, and the unplugging the vacuum line at the distro definitely changes RPMs.

Babak

0.1V at idle is not a problem since there is no load on the engine and impossible to do any damage. In other words, it's nothing to worry about. Also, your voltmeter may not be reading such a small voltage accurately without interfering with the sensor signal anyway.

I plugged a copper wire into the 02 to wiring harness receptacle. Then you can have a reading with the system hooked up.

Like Joe Bob says, if at idle you are at 0.1v AND the O2 sensor is warmed up that concerns me that your baseline 3mm allen wrench (CO) setting is too lean. Baseline means like the older CIS systems with no Lambda system to richen up lean mixtures at idle - or lean out rich settings.

When you rev it up the O2 system is disabled (I think beyond 35% throttle) and your WOT system dumps a robust amount of fuel in the system. That may explain the high reading.

As mentioned, drop (or steal) the $40-$50 at Autozone for an analog Actron dwell meter and then you can read what your frequency valve is doing.

Try the first thing I mentioned and get the O2 reading when it is functioning. If it reads in the 0.45 volt range and goes up and down a tenth volt your O2 system is compensating for a lean baseline.

Also search on psalt and read some of his discussions. He know lots about the detailed functioning of the Lambda system when at idle, off idle and open throttle.

Agree, but after the O2 system shuts off and before the WOT switch kicks in can there be an undesirable lean passage at higher RPMs?

I do not know if the CO setting will impact the WOT air fuel ratio.

How would you guys interpret the 0.1V at idle?

Babak,

Meaningless.

The way all closed loop engine management systems work is to dither the mixture rich and lean around stoich, to make an average 14.7:1 ratio for the converter. Steady state readings on a narrow band, unheated O2 sensor are not very accurate. About all you can say is it may be lean of stoich, but it could mean a misfire, or the sensor temp is too low, or the sensor is dirty or worn out, or several other things. A narrow band sensor is more like a switch, accurate around stoich with a steeply sloping curve that is not suitable for steady state measurements. It was never designed as a meter, you need a wide band heated sensor to do that.

The way to set the mixture on an 82 SC is with a dwell meter. This will show you how much the system has to change the pulsewidth of the FV to correct the mixture back to stoich and the response time of the O2 sensor. It will also show you if the system has the correct default duty cycles outside closed loop. Everything you need to know can be found on this site.

Too many fail safes on a Bosch CIS injection system which always defaults to rich to be worried about that.

Unfortunately, that is not true for a 82 SC with lambda CIS. When the fuse, relay, ECU or FV fails, the system default is far too lean, especially at WOT. When I first learned this 20 years ago, I thought, what is up with those dumb BOSCH engineers ? But it was the method they chose to get a wider range of mixture trim. In CIS lambda, the K Basic part of the system is calibrated way too lean. When the FV is closed, because it failed or the ECU is not powered, no lower chamber pressure can be bled off and the amount of fuel delivered to the injectors is lower. The FV must be pulsing at the default duty cycle for the mixture to be anywhere near rich enough. An 82 SC with a blown fuse or relay will immediate run like crap because it is too lean everywhere on the fuel curve. There is no "limp home" rich default backup like some EFI systems.

"How would you guys interpret the 0.1V at idle?"

psalt: "Meaningless."

And then this:

psalt: "About all you can say is it may be lean of stoich, but it could mean a misfire, or the sensor temp is too low, or the sensor is dirty or worn out, or several other things."

Statements are contradictory!

Actually, the second statement is correct and states what others have basically said
in this thread. But like most Pelican posts, many like to re-state what others have
stated or what has been posted elsewhere.

Yawn.

Same acidic and counter-productive comments from a sharp but snarkey dude. Happy New Year, Loren.

Thanks guys - I tried oxygen sensor on/off, various idle settings, etc.. but was still getting backfires when the engine was warm (not hot). I finally asked John at Valhalla check the mixture. It was way off! Timing was quite close though. He adjusted it and now I'm motoring quite finely.

Thanks all (especially John!),

Babak