O2 sensor idea

[Steve Gorkowski]

Yes you can but remember after you hit 1.2v out it folds back to near 0v.

Steve

Greg Hermann wrote:

> >On Mon, 19 Apr 1999, Greg Hermann wrote:
> >
> >> >On Mon, 19 Apr 1999, Ken Kelly wrote:
> >> >
> >> >> Roger,
> >> >>
> >> >> The GM PCM only looks at crossings of 14.7 AFM, that is why
> >> >> they are only calibrated at that point. As soon as you go to
> >> >> Wide open throttle the PCM ignores the O2 sensor. If the O2
> >> >> sensor was linear from 14.7 to 12.0 AFR, You could rewrite
> >> >> the software, so it would stay in closed loop, but it would
> >> >> be a major rewrite.
> >> >>
> >> >>           Ken
> >> >
> >> >I know it works that way.  I just want an accurate O2 sensor so when I
> >> >am burning new chips I actaully know the O2 sensor is reliable.  The
> >> >diacom output should have accurate O2 readings, and then tuning would
> >> >be easier.  I have to be careful right now becuase I really don't know
> >> >what voltage is exactly 12.5:1, I am just guessing.
> >>
> >> Guessing is the right word for it, Roger. and with anything less than a
> >> well calibrated wide band O2 unit, you would be continuing to guess. But I
> >> think you knew that!
> >>
> >
> >That is why I want to get a well calibrated wide band unit and setup a
> >pic or something similar to give the kind of results the stock
> >computer wants, and give mv readings that actually mean something.
>
> As I said earlier, Roger, it MAY (not sure yet) be possible to bias a wide
> band O2 sensor with a constant ion pumping current so that it crosses at
> any A/F ratio you want, in the same way that a standard EGO or HEGO sensor
> does--Changing the bias say for cruise or WOT, and leaving an oem ecm in
> closed loop would allow a programmable selection of A/F ratios  for
> different operating conditions. (The bias adjustment is what would fool the
> ecu, the sensor would act like normal EGO sensor at 14.7:1.) The bias and
> its changes would need to be dealt with outside of the oem ecu.
>
> Regards, Greg
> >
> >                       Roger