[Diy_efi] DIY-WB -- Rcal vs. Ip samples

[Garfield Willis]

On Tue, 22 Oct 2002 07:45:57 +1000, Peter Gargano
<peter at techedge.com.au> wrote:

>As the sensor ages, you can expect some variation from the initial =
free-air
>Ip, so any kind of serious tuning use should be periodically checking =
(and
>adjusting) the free-air Vout to be 1.50 Volts above the Vstoic value.

This is completely bogus. It will NOT result in a free-air calibration
of any sort, nor compensation for aging. You can't just pick some
arbitrary value of Ip + some downstream gain, and say "4.00V out" is
where free air should indicate. Like I pointed out before, you're just
chasing your tail.

Where's you calibration curve that says what nominal or mean Ip is for
free-air?, you basing it on your massive sample of 11 data points? or on
a curve copied from some *other* type of sensor? If you don't believe
that the curve for the Honda/NTK is different than the one in the SAE
article, for christ sake, LOOK at the Ip data for free-air you just
posted. THEN look at the graph (Fig. 9) shown in that SAE paper. Do the
free-air Ip values look the same? NO, of course they don't. Because
they're NOT the same sensors!

That's how both an actual calibration curve AND how NTK based their
selection of the Honda CalR values works. You first have to know the
distribution of Ip vrs AFR curves over a wide enough sample of sensors,
before you can say anything about free-air or any other kind of
calibration. In short, you HAVE no calibration at this point. And until
you have the actual Ip distribution and ALSO a distribution of the full
sweep curve of Ip vrs AFR, you can't even begin to fathom how the
factory Honda/NTK CalR algorithm even WORKS.

I've generously disclosed that the CalRs were lumpy, that the curve
wasn't the same as the SAE paper's curve, that the DIY-WB circuitry was
flawed in both the CalR, the heater, and the Ip control, and would
result in major static AND transient errors, in the whole-AFRs range of
errors. Slowly people should be coming around to the realization that
those statements weren't merely partisan, but simply factual technical
comments. But what do you do with them? Like the proverbial
'unteachable' children, you eat your books! Now we have this latest
"adjustment to some arbitrary point" =3D=3D 'free-air calibration'. NOT.
There's no excuse for such an obvious error; it's just a lazy (if not
outright fraudulent) attempt to blow smoke over the issues and hope
nobody notices.

Mere rote copying of someone else's work NEVER leads to understanding,
that's why it's outlawed in school; and especially when you've copied
someone else who also didn't have any understanding, you're doubley
handicapped.

You will NEVER fix the numerous DIY-WB flaws until you begin to
understand the topic. And judging from the fact you guys haven't even
begun to study even the rudiments yet, I don't think anyone should hold
their breath.

Gar


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi