Wide Ratio O2

[Steven Gorkowski]

Gar

I think you have a good handle on the wide ratio project and will be
waiting for your results . I have a person to test the O2 meter curves .
My goal is to make a very cheap (under $100) meter, very low parts count
and a serial data stream and linear 0 to 5v out option.

Steve

garfield at pilgrimhouse.com wrote:

> On Fri, 27 Mar 1998 22:15:18 -0600, Steven Gorkowski <kb4mxo at mwt.net>
> wrote:
>
> >Thanks for this information I needed the graphs to make a standard O2
>
> >sensor work and you and this company gave me this information. I have
>
> >used a standard O2 sensor in the past but without temp regulation and
>
> >temp sensing and agree that it is not as accurate as I need but if
> the
> >uC(pic) can regulate the heater so it is close to a temp curve . I
> can
> >build a meter for under $100 and will output A/f ratio to a standard
> >rs232 port.
>
> Hey Steve.
>
> Say, couple things I just thot of. First, them SplitSecond guys
> claimed
> the sensor they sell (graph shown) is a somewhat "wider-range" (for a
> cheap heated sensor) than usual, that's made by NTK! You might wanna
> sniff around with them about that. If you made it clear that you
> weren't
> interested in their box, but only their sensor, and wanted to know a
> second source for it, they just "might" be willing to share that info.
>
> Maybe sell you one sensor (I think their price is $60 or so), with the
>
> understanding that they'd give you the automotive source if you needed
>
> to second-source it. Point is I'm not sure those curves are
> representative of any ole heated O2 sensor. Might wanna check on that.
>
> Second, you bring up a good point, in that if you used the "sense the
> sensor's resistance" technique like Peter has illustrated, and just
> held
> it fairly constant by controlling the heater current (meaning the
> element's temp was fairly constant), then FOR THAT SENSOR, if you
> could
> run it *just once* in league with a previously calibrated O2 Meter and
>
> calibrate your own Vs versus AFR, even if you didn't know the
> temperature exactly, nor had the temp. calibration curves, you could
> still get the thing calibrated decent and repeatable enough. You don't
>
> necessarily need to know the temp. precisely if you have some other
> AFR
> calibration standard to use, as long as you know you can control/keep
> the sensor at that calibration temp. from then on, after the
> calibration
> run is over.
>
> I hate to see you go to all this trouble, though. Maybe if you can
> wait
> a couple more weeks, we'll know if these NTK UEGO sensors are gonna
> work
> out. Yeah, the sensor is more moolah ($120 vrs. $25-60), but if they
> work well, the electronics will be not much more than a few bucks on
> top
> of that, and the thing might be self-calibrating (free air std),
> taking
> away the need to borrow some other standard against which to
> calibrate.
> But hey, if you can make the calibration & heater control of a
> ordinary
> heated O2 sensor give 0.5 AFR accuracy, maybe we build one snazzy $200
>
> one that's self-calibrating, and then loan it around to calibrate the
> <$100 ones. Hee hee. Kinda a pyramid scheme.
>
> Either way, though, it's looking much more positive that we'll get
> something, one way or another, to give us some better O2/mixture/AFR
> readings. This has been a nice productive thread so far; lots of good
> &
> diverse ideas, dudes.
>
> Gar