Wide Band Oxygen Sensor Web Page

[Bernd Felsche]

I wrote:

> Paul tapped away at the keyboard with:
> > For those who couldn't access my web page, you know that useless piece of
> > disinformation, the
> > ip address is;

> > 196.22.202.162

> > The problems seem to be from those of you in the UK.
> 
> Useless information as usual. :-)

> Looks like your domain name isn't known outside certain nameserver
> authority spaces.

> The schematic shown is quite a bit more detailed than what Bosch
> publish in the latest edition of their "green book".

Oh; I also meant to add that voltage-regulation (and current-
limiting) circuits are a dime a dozen. 

Just a thought or two as it's nearing midnight and the moon is out:

If one is integrating the sensor into an engine-management system,
much of the circuitry can be "replaced" by software; the main
feedback loop using an ADC channel (or analogue comparator) and the
current "drive" control by a DAC. The Ip would result from an output
of the ECU and hence wouldn't need to be measured - though one might
like to ensure that certain, safe limits can never be exceeded,
perhaps with a digital interrupt to the ECU to tell it that the pump
current is too high, for example.

Oddly-enough, well maybe not, I think all that digital control of
the sensor could be done with a simple 8-bit MPU - so if you wanted
to, you could *automatically*calibrate* the sensor by setting EEPROM
values in a matched MPU. An 8-pin device such as an ATtiny12 is
sufficient if you use an external register for a DAC and two bits to
control up-down counting. OTOH, a chip with more digital outputs could
control the current source directly, reducing the parts count.

The really great benefit is that you could make the sensor look like
a switching O2 sensor without great difficulty and trivial external
circuitry.

Interfacing to a more advanced ECU could be via in-built SPI.

It should go without saying that the "digital sensor" would also be
able to ensure that the actual sensor is treated well during warmup
and hot-starting. Returning "sensible" results the the ECU as
appropriate.

You could get really sophisticated and measure the temperature of
the sensor and exhaust gas for example, to get some idea of the
validity (reliability) of the readings. (Did anybody else get the
idea from the reading of the patent, that the cell needs to be at a
temperature greater than the exhaust gas?)

-- 
Bernd Felsche - Innovative Reckoning
Perth, Western Australia
----------------------------------------------------------------------------
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