With the O2 sensor disconnected, the ECU may have been sensing noise on the the input, particularly if that input has a high impedance (which is likely).
From ChatGPT: For most zirconia-type oxygen sensors, the impedance can be quite high, often in the range of several megaohms (MΩ). Specifically, it can range from around 1 MΩ to 10 MΩ or more, depending on the operating temperature and the specific design of the sensor.
So the ECU’s input would be a high impedance as well, and l believe we’ve found a probable cause.
@1998_K1500_Sub, Good call!
You are definitely onto something. First off, elsewhere in the FSM there is a recommendation to the
tech to employ a multimeter with a 20 Megohm input resistance. Which would explain the note in the
FSM about seeing .32v (vs the expected midpoint of .45v) if using the older/lower 10 Megohm resistance
multimeter.
****
It's an oversimplification, but to normal, well-adjusted folks I would describe the ECM/PCM/VCM O2
sensor circuitry that's monitoring the output of the O2 sensor as a balanced teeter/totter, with the
provided 0.45v bias as the center pivot point. And instead of the very low internal resistance of a
car battery, generating a difference voltage from the presence of oxygen on only 1 side (~900mV) vs
both sides (~100mV) of the zirconia barrier is, by the inner workings at the very center of the
narrowband O2 sensor, a very high internal resistance voltage source.
Given this, the very design of the O2 sensor is such that in order for this very weak voltage source to be
able to 'push/pull' the monitoring circuit in the computer, then that circuit must match the impedance (actually varying
DC voltage at up to a rate of 6 Hz depending upon speed of overall feedback loop) of the sensor it's monitoring.
(See 1st attachment)
****
As for the computer seeing RFI on this high impedance circuit and getting dazed & confused by the
spurious signalling, your theory answers the questions I had last night. If I went back to the engineers
that coded the onboard diagnostics and asked why it generated a P0430 with the disconnected wire, they
would stop me right away and say that their tests are written with assumed input conditions, and the
situation of having a disconnected wire is an
illegal physical condition in the circuit being monitored, and
therefore not meeting the input conditions they wrote their diags to work within.
In English, having a disconnected sensor with the wire in the harness turned it into an antenna on a
circuit with high impedance would certainly couple the RFI-rich engine bay chaos into the computer. And
as computer system diagnostic writers would explain to me, the results of this scenario would simply
be 'unpredictable'. (Back when I actually did this for a paycheck, I would escalate stuff like this up to
corporate, and much brighter bulbs than I would respond back with phrases like 'illegal condition' leading
to 'unpredictable outcomes'. (BTW, I learned so much from their architectural approach to how failures
would counterintuitively present on the customer site in the field.) Some of those guys were so sharp
that it was humbling to read their thought process. (!)
By the way, if you are curious about what was superimposed onto the disconnected O2 sensor wire
connected to the computer, you don't need a $$$$ pico scope to get an exact idea. Just grab a small
portable AM radio, tune it between stations, start the engine, hold the radio in the general area of
the O2 sensor, and by listening you will get a feel of what the computer was trying to make sense
of/develop a response to the strange closed-loop feed back?
NOTE: Most AM radios are directional receivers due to their reliance on ferrite rod antennas. (Watch
an old timer, and without even thinking they will orient their small transistor radio in order to get
the best reception.) If you have 7 good wires and 1 is failing the nighttime firefly test, grab an AM
radio and see if you can hear the difference between the good cylinders vs the 1 outlier. Using the
directional feature of the AM radio, a lot of times you can home in on the extraneous arc. (Normally
the external arcing is louder than what's going on inside the distributor and inside the
combustion chambers.)
...But I digress.
This is already longer than I wanted it to be. But for awhile now I've had an interest
in interpreting the O2 sensor signalling as seen in stored graphs in order to glean valid broke/fix
info from them. But since this is too long, instead of discussing the last 3 attachments I'll simply
put them there and leave them as an exercise for the individual(s) still reading this. :0)
****
And the give & take in this thread is *exactly* why I decided to join this forum. More heat than light
is shed on subjects like this, and I enjoy learning as much as I can when it comes to troubleshooting
our GMT400 powerplants.
Cheers --
