2013 Tahoe 5.3 P015D P015B pre-cat O2 sensor delayed response

Hello Everyone,

I am temporarily working as an apprentice technician at a GM dealership while school is out. I am working under another technician who has the last say in diagnosis and calling parts. He really wants to throw O2 sensors at this car, but I'm not convinced because we haven't done any test that definitively proves to me that they're bad. Maybe we have, and I just don't understand the data we collected. It came in late and there wasn't much time for testing.

Here's the situation:
The car came in with historic codes. The owner said that the MIL was coming on intermittently after our on-site body shop did some work to the truck, but I don't yet know what exactly that work was. The car apparently had the MIL on when it was handed over to the service advisor, but it was not illuminated by the time it got to our hands. Freeze frame data showed that it happened at about 1100rpm and 15% engine load. I hooked up a tech 2 and started plotting the pre-cat O2 sensor voltages at idle by themselves so the data would update quickly. I saw a normal looking o2 signal, but the signal stayed high for a while before dropping lean again, which made the frequency about 0.33Hz, or one every 3 seconds or so. It was between 50mV and 900mV. The drop from 600mV to 300mV was always fast, I never saw anything that would indicate a slow transition to me. We introduced a vacuum leak past the throttlebody by undoing a hose, and immediately saw the signals go lean, then start oscillating again when fuel trim adjusted to match. WOT on a test drive showed rich on both sensors. Fuel trims were always negligible and stayed around 0%. The truck does not seem to have any exhaust leaks.

I was away for a moment when the service manager asked for an update. The technician I work with mentioned that we haven't found anything definitive, and the code hasn't shown up, so we may opt to return the car until the MIL is more frequent. He was told that was not acceptable, and that we need to try harder to find the issue.

I decided to hook my picoscope 2205a to the O2 sensor signal at the ecm with a T-pin. I wanted to measure the drop and rise time of both sensor signals between 600 and 300mV to make sure it was lower than 100ms. Here's where I ran into an issue. All I got from (what I think is) the signal wire was noise. (Pin x2-59, right?) My scope ground was good and read battery positive and the heater control square wave (pin x2-13) just fine. I used my multimeter to make sure there was continuity between the t pin and the pin of the connector for each test. The tech 2 shows a sane-ish signal, so I know that I'm doing something wrong.

I know a cold o2 sensor or a bad heater will cause it to respond slowly, but I was under the vague impression that the heater is almost irrelevant when the car is warm. Should I do heater tests? I definitely need to visit chapter 5 material in more depth.

What am I doing wrong? What would you guys suggest doing next? One tech said that the long high voltage times may be due to a "lazy o2 sensor that just needs to be replaced", but I don't really buy that without a test to definitively show it. `

Ethanol content in the fuel was 7% according to our test kit. One guy on a forum said using ethanol free fuel will cause these codes to pop up for him, but there was no data or reasoning, so I don't really buy that either.

The tech I work with was 100% ready to throw o2's at this today, but I asked for more time to look at this in the morning. Hopefully I can avoid being a parts changer tomorrow.

Any service bulletins or pcm updates?? Also you could swap the o2 sensor from bank to bank.

No bulletins or campaigns. We are told not to update the ECM unless a bulletin, recall, part replacement procedure, or similar instructs us to.

Both banks are showing similar behavior, so I didn't think swapping them would do anything. Is there something specific you would expect to see by doing that?

Disregard what I said about signal frequency. the x axis on the tech 2 was frames, not seconds. There is no way of telling what the frequency is without a scope, but I still have the noise issue to figure out before I can do that.

Looks like you need to zoom in on the signal waveform.

I went all the way down to about 50ms/div and up to 5sec/div while trying everything in between. The signal always looked like high frequency noise.

The noise ended up being from a loose banana jack adapter on my probe. I was able to scope the signal with the 12 bit simulated resolution enhancement and found it to be slow. I've attached an image of what I found.

Good to hear you got confirmation of bad o2 as well as diagnosis on the banana jacks. Its a good sign when you catch your equipment having issues before going down a rabbit hole with a car. Bravo

I got burned on this with a Mercedes a couple years ago. Turned out the slow response codes were caused by a cracked catalytic converter. No exhaust noise but was leaking enough air in to set slow response code. Unfortunately, I'd replaced the O2's before I found the cracked exhaust. Live and learn.

When looking at standard O2 sensor response I now use a cheap OBD2 ELM327 dongle with the InCarDoc Pro Android app I found the nice clear graphing it gives is more useful than my Autel and Launch tools.

This is a screen cap from InCarDoc that I annotated