[Diy_efi] O2 says rich, car runs lean..

[Tom Visel]

Look at your oxygen sensor voltage, preferably with an oscilloscope, 
though even a scan tool or a high-impedance DVOM would do, as long as it 
doesn't do too bad a job of "smothering" away the sensor's natural 
jitters with its averaging function.  Some meters that have a bar graph 
at the bottom will let the bar graph twitch while the numbers get 
averaged out to something smooth and useless.  A bar-graph ricer gauge 
will do also; the cheaper ones are better for this purpose because they 
let the original signal show through without cleaning it up for prime time.

What you are looking for is movement of the sensor voltage - is it 
jittering rich/lean or rich/richer, or is it parked at some voltage?  If 
it is parked at some rich (>450 mV) reading and staying there, odds are 
the sensor is dead.  If it is reading over 1 volt, odds are you have a 
bad ground for the sensor, and stray electrons are making their way home 
on the oxygen sensor line.  This is more prevalent on vehicles with 
headers, due to the gasket trickery that often makes for a less than 
perfect ground at the flanges.  Heated oxygen sensors have even more 
problems like this, because the 12 volts supplied to the heater needs 
its own GOOD ground, and it will use whatever other wire is available if 
the heater ground wire is bad.  This is more of a problem on fresh 
custom builds and vehicles with older sensors.

To quick-test your oxygen sensor ground, unplug the alternator while 
watching the O2S output.  Did it improve?  If so, bad ground.  More 
electrical load = more strain on grounds.  Another method of testing is 
to run a jumper cable from the battery negative post to the housing of 
the O2 sensor to act as an extra ground.  If that improves the signal 
quality, you know your grounds need work.

BTW, your oxygen sensor can be switching rich/lean at the sensor - going 
from .25 to .75 volts, say - and a bad ground will turn that into a 
signal that varies from .6 to 1.1 volts, which the ECM would interpret 
as all rich, all the time.

HTH!
TomV

David Rowley wrote:

>Hi all,
>   
>  I have a chevy 383 motor with Vortec Heads, GMPP hot cam (218-228 duration, roller), Edelbrock RPM Air gap manifold and a Holley 670 throttle body injection unit.
>   
>  I have a problem where the car runs and idles great, until it goes to closed loop mode.  When this happens, the ECM starts leaning out the AFR (the block learn goes down the minimum, 108 in this case) to the point where it will barely run.  Why is the O2 seeing a rich mixture when it appears it is not?  I have tried two O2 sensors with the same result.  Does it have to do the the cam duration and manifold type?
>   
>  I have tuned a few vehicles and I have always been able to rely on the O2 sensor ( with the Integrator and BLM numbers) to get them to run right, until now.  
>   
>  Has anyone out there seen anything like this?  
>   
>  Thanks,
>   
>  Dave
>   
>
> __________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://lists.diy-efi.org/mailman/listinfo/diy_efi
>
>
>  
>