AFR calc from Diacom data... was RE: Adding another O2 sensor

[Walter Sherwin]

A few thought starters............

>
> Need to get the high speed link going.  Has anyone encountered wiring
> problems causing differing O2 readings?


Yes!  You should not alter the wire lead affixed to the O2 sensor in any way
(ie: to lengthen it for relocation into a header collector/tube).
Similarily, if modifying the harness wiring upstream of the O2 connector
area, make sure that you do not solder the terminal onto the wire and that
you leave the insulation back a bit from the end of the crimped-on terminal.
As strange as this sounds, some but not all, O2 sensors by design draw their
reference air from between the inner side of the insulation and the copper
conductor strands.  If you impede this path, then these types of sensors may
act funny.  It is virtually impossible to look at a sensor and determine
exactly what flavour it is, so that's why you'll find warnings to this
effect in many of the GM Service Manuals and Emissions Guides.

Also, never tie another electronic device into the O2 sensor circuit.  This
particular circuitry is greatly susceptible to impedance loading issues and
you will get erroneous O2 readings as a result.  I've seen low-buck A/F
meters skew the reading by as much as 300 mv when "T'd" into the ECM O2
loop.


I'm considering running a dedicated
> ground from the O2 case to the ECM.  Bad idea?


Excellent idea, but probably best accomplished in a slightly different
manner, especially since you mentioned that you have headers and custom
exhaust.  Here's the deal........In order for your ECM to be capable of
accurately measuring the small O2 sensor voltages reliably, it must be
electrically connected to both the O2 signal (O2 Hi Sense) and the ground
plane above which that signal was produced (O2 Lo Sense).  In typical GM
fashion, the older applications tended to connect the "O2 Lo Sense"
circuitry of the ECM to the engine block via one of the bolts at the
thermostat housing, and in turn connected the O2 signal wire directly to the
ECM.  So,  it was ***assumed*** that the shell of the zirc O2 sensor would
ground to the bulky cast exhaust manifolds, and then to the bolts attaching
it, and then to..........yadayadayada..........up to the star terminal
connected to a wire which in turn connected back to the ECM.  Well,  lets
just say that this is all well and good if there's no rust, sealants, years
of neglect, or custom exhaust gaskets and coatings in the way.

Refer to several recent threads on this list regarding single wire heated
and unheated O2 sensors equipped with redundant ground wires (many specific
P/N's were published).  In situations where the O2 signal ground may be
"suspect", the redundant lead can be connected back to the same point at
which the O2 Lo Sense circuit connects to the thermostat housing via the
star washer.  This is a bit easier that fishing a wire all the way back to
the ECM case if the case happens to be in the cabin, and actually carries
with it a few other minor benefits as well.

Alternately,  you can retain your stock O2, and then electricaly jumper your
exhaust headers back to the engine block.  However,  you hadn't better have
any HPC coating (or things similar) between the threads of the O2 shell and
your O2 boss.

Actually, on a related note,  I usually advocate the introduction of a few
selected ground plane jumpers in the applications I work on.  One of those
is usualy from the thermostat housing stud over to a clean spot on the
engine block.  The other is usually from this same point back to battery
negative.  Cheap insurance.


>
> Thanks a bunch!  I'll keep working on it.
> Steve (starting wideband research effort)
>


Happy Hunting
Walt.


----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org