O2 sensor question - how it works

[Garfield Willis]

On Fri, 24 Mar 2000 22:46:02 -0600, "Brian Franchuk" <franchuk at mm.com>
wrote:

>For a good explanation of how the LAF oxygen sensor works see 
>the following web page:
>
>	http://www.tech2tech.net/training/laf.htm

Oh my, well, that's a noble attempt, and lots of basic things mentioned
are correct, but there's a bunch of stuff that's also NOT correct in
this discourse. Suffice it to say, the secret salsa recipe is safe, from
anyone who thinks this is actually the way the sensor works from an
operational standpoint. He's got the basic physics of the thing right,
but some of the operational things are completely off. The wire/terminal
he refers to as a "reference" is infact a floating common for both the
nernst and pump cells, and swings all over the place. Testing it for
2.7V to confirm correct function is pretty bogus. Secondly, no Honda-NTK
sensor electrics controls the pump voltage, as he describes it. It
cannot work that way, because the pump cell's impedance also moves all
over the place even for any given AFR, sometimes making large
excursions, as the pump cell finally achieves balance of O2 ion levels.
The pump *current* mirrors the AFR, not the pump voltage; that can vary
widely as the impedance of the pump itself changes. You learn this when
you work with these things and watch their dynamics. He might not have
seen this in steady-state, but to say it's a "reference" with some
relatively constant value, misconstrues the workings of the sensor. Best
to consult the SAE/NTK paper if you want to understand this better. Last
thing I'm gonna mention, cuz it's potentially important for people who
will use EGOR and go about buying their own sensor, is he says the
calibration resistor value "is typically 4.3K ohms". This is potentially
very misleading. Perhaps he's seen a few, and based his conclusion on
this, but I've measured quite a few of them, and the cal resistor on
these can vary from as low as 1.5K to as high as 10-15K. It's used as a
factory-set "parallel trim value" in the ECU circuitry (which doesn't
have the luxury of doing a free-air calibration like we proformance
users can), and as such, varys widely for only slight (15% overall)
changes in the sensor calibration itself. But there's nothing "typical"
about the cal resistor value. It varys widely.

Not to carp (much) at what would otherwise be a very useful beginning
tutorial on these interesting devices. I just wish the guy had looked
into the NTK/SAE tech docs so he could combine his practicals with some
understanding. Auto mechanics and diagnosticians would do well to
attempt understanding the underlying device from the source, rather than
making up some terminal measurement rules on what to expect, based on
their experience or factory training alone. Oh well, what's a mother to
do. :)  It's sorta close, but no cigar. A little help from the NTK SAE
paper (#920234), and he could have avoided some of the more glaring
mistakes.

Gar


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