Date: Thu, 3 Jul 1997 14:32:26 -0700 (PDT)
Sender: Vanagon Mailing List <vanagon@vanagon.com>
From: Tom Young <young@sherlock.sims.berkeley.edu>
Subject: Re: Oxygen Sensors
On Thu, 3 Jul 1997, Craig Rogers wrote:
> I'm wondering if there is a way to check an O2 sensor to see if it is
> operating properly.
Yes there is, and the Bosch FI book tells how. Called the "Lean Stop
Test" and the "Rich Stop Test", they require a CO meter. However, if the
CO doesn't return to the original value (for the two tests) it then says
"...check the fuel pressure regulator and system wiring. If the
regulator and wiring are not at fault, check the lambda sensor by
measuring its voltage output with the engine running at 3000 rpm. If
there is no voltage fluctuation (in millivolts), replace the sensor."
> It seems as though there would be a lot of variables,
> but I think that the ECU supplies the sensor with around 500mV and the
> resistance of the sensor varies, depending on the temperature of the
> exhaust.
No, the O2 sensor is a form of battery, and sends out voltage in response
to the difference between oxygen in the exhaust and ambient oxygen.
Exhaust temperature is only important in the case of the single wire
sensors, since they have to be brought up to operating temperatures to work.
> Since the output of the ECU is current limited(I haven't checked
> the value), the voltage at the output varies and this is what the ECU
> actually measures. So a particular CO level should correspond to a
> particular voltage at the ECU.
I'm not sure what you're saying here exactly, but you're right, the
output of the O2 sensor varies with different levels of "richness" (and,
hence, CO) so you can use the O2 sensor and a voltmeter to set your idle
fuel mixture.
> If we could test the CO and check the
> voltage on a number of vehicles and make a chart, then if the numbers
> correspond, one could tell if it's the sensor or some other part of the
> system that is functioning incorrectly. If the numbers are right, then
> it's not the O2 sensor.
I don't think this is right. Before I fixed my air flow sensor, the
output from my O2 sensor was "pegged" right around .9 volts no matter how
I adjusted the idle fuel mixture. The O2 sensor was telling me my
mixture was too rich (and, I suppose, that my CO was too high), but how
would this tell me if the O2 sensor was faulty? Once I fixed my air flow
sensor I was able to adjust the idle fuel mixture, and see the results
show up on the voltmeter connected to the O2 sensor. I'm not sure how O2
sensors fail (although, I'd guess, like most batteries, they stop
delivering current) but if it was possible for an O2 sensor to fail
"rich" (i.e., put out a constant high voltage, no matter what the
mixture) then my initial (high voltage) readings could have come from a
bad O2 sensor, OR from some other problem that was causing a too-rich
mixture.
>[snip]
> This could also work in reverse. If the system is
> functioning correctly, the CO level could be checked by its corresponding
> voltage at the O2 sensor and a $20 voltmeter could be used to adjust the
> CO. That would be a lot cheaper than other alternatives. Has this been
> done? Are there other ways? Anyone?
Yes, you can set your idle fuel mixture properly (and get the CO level
right) by using the O2 meter and a voltmeter.
---------------------------------------------------------------------
Tom Young young@sherlock.SIMS.Berkeley.EDU
Lafayette, CA 94549 '81 Vanagon
---------------------------------------------------------------------