Yes, and the new ones all tested okay, both the generic AB one and the Bosch.

Unfortunately, I don't think we've solved our issue.

We're getting the P0134 again, and it is persistent.

The confusing thing is that we are getting a reading (albeit low voltage) from the sensor initially at idle, and the it drops out and shows 0.0v. I took it for a 30 mile test drive and it never moved from 0.0v, while the other front sensor showed varying values in the 0-1v range.

I took a video of the idle and that bank only read 0.02 and 0.03 volts, while the other bank ranged from 0.03 to 0.79 and many values in between.

It acts like one of those oddball things where there is some component that is impacted by heat, like a solder joint or electronic component.

I have a fan blowing on the engine now, and will check it when it is cooled down. If it shows some amount of voltage and then it goes away once everything warms up again, at least I have a clue.

The protective wire loom on the drivers side was badly dry rotted, and came apart. Does this look typical where the red and orange wires go into some heat shrink, or am I possibly looking at a repair splice?

 

Maybe cut that open and check for a bad splice. At least compare to the other side.

Voltage should rise fairly quickly to around 1 volt (rich condition) when cold started, both sides. Don't know if the heater circuits are de-energized after the engine heats up.

 

Fun times.

I'll take a closer look at that heat shrink looking section once it cools off again. Any insight if that heat shrink area looking area looks normal? Or does it look like a splice.

More data points:

So, I woke up today to a cold engine (40-ish degrees outside overnight), turned the ignition to on, plugged in the Nanocom, both upstream bank A and bank B sensors read 0.44v.

I started the engine and over the course of two minutes the bank A sensor slowly and progressively ticked down 0.44, 0.43, 0.42 etc until it got to 0.04 and bounced back and forth between 0.04 and 0.05.

Meanwhile the bank B sensor ranged from 0.78 to 0.03 and oscillated up and down across this range.

After a few minutes, I took a 5 mile drive. After about 2 miles, the bank A sensor went to 0.00. It popped a reading of 0.05 for a second and then went to 0.00 and stayed there for the rest of the loop.

Meanwhile, bank B continued to fluctuate in the same 0.78 to 0.03 ish range.

 

So we know bank B is working well. Bank A is not working well, but we don't know why. So, previous advice to swap sensors makes sense in order to find out where the problem is.

The B side sensor is known to work well, so if it is installed on the A side and does not work anymore, then there is a problem on that side, not the sensor, for example a wiring splice issue or ECM issue. Likewise, if the B sensor works well on the A side, we know that sensor is good. Of course taking care not to damage the sensors during removal and installation.

Each bank's fuel trims are controlled independently by the ECM. Changing the short term fuel trim is how the ECM changes between rich and lean conditions, so the O2 voltages should modulate between near 0.1 (lean) and 1.0 (rich), when the engine is warmed up (closed loop mode). The ECM looks for when the voltage crosses the middle voltage (stoichiometric value, around 0.45 volts) to make adjustments to maintain optimal conditions for efficient combustion. Consequently, monitoring both short term and long term fuel trims can also provide data indicating proper operation of the engine. Have you monitored those values?

 

Thank you.

I have not monitored STFT or LTFT on this vehicle (but have on other vehicles). I can do that.

I worry that regardless of fuel trim data, I am definitely not getting a reading from the bank A upstream sensor, and need to rectify that for any other stuff to make sense.

I asked my son to swap upstream sensors today.

 

Okay, the bizzare continues.

My son swapped upstream sensors and the "good" sensor moved from bank B to bank A. Bank A had readings that fluctuated appropriately cycling from high to low. Bank B quickly went to 0.00 volts and stayed there.

We figured that this meant we had a bad O2 sensor (was in bank A, now in bank B).

So we installed another sensor that passed the propane/multimeter test into bank B, thinking we had it sorted.

Wrong. As soon as we put a suspected working sensor in bank B, both banks read erratically. Bank A quickly went to 0.00 and bank B followed suit reading 0.00. Frack!

So we tried another suspected working sensor in bank B. That one also had low and slow moving readings on both bank A and bank B, both eventually going to 0.01 and 0.00. Dang it.

How is it possible to get good readings on a sensor in bank B, swap it then also give good readings on bank A, but then put another sensor in bank B and everything go haywire?

Do the banks have some interdependency?

At this point, our next move feels like to try yet another new sensor in bank B. The one currently in bank A should be the "good" one that most recently read correctly.

If anyone can explain this bizarre behavior, we'd be grateful.

This is incredibly frustrating. I've never experienced an issue like this before.

 

They are not related. They operate completely independently. If you look at this video you can see how they should operate. You can have other issues that cause it to go to zero (it is actually lean on that side, usually injector o-rings). The sure fire test is fire it up stone cold, monitor the voltage using an analog data logger like the torque app. Voltage should start out close to 1 volt. If it does the sensor is good. If it goes to zero after that you have a vacuum leak.

 

Frustrating, I know, but you are actually making progress. STFT large positive value (up to +25% max) could be another indicator of a vacuum leak, as the ECM tries to add more fuel to compensate for the added oxygen. STFT is reduced if the O2 sensor starts to read rich (near 1 volt).

It is super easy to get messed up if one or both of the O2 sensors are not good. I know from experience with my trucks. I personally don't like to use a propane torch on the sensors because that could also cause damage if overheated, or heated too rapidly.

 

Dead sensor on bank 2?

OBD Fusion



Given the description above, I could have easily overheated the new sensor with the propane test and damaged them.

And I have smoke tested this engine several times and fixed all vacuum leaks. I can test again tonight, but I will be shocked if we have a vacuum leak.

Also, a vacuum leak seems inconsistent with how the same sensor had correct readings in bank B and also had good readings when swapped to bank A.

It feels more likely that I have one good upstream O2 sensor, currently installed in bank A, and a faulty sensor installed in bank B.

I need someone to check my logic, please.

 

Would an injector o-ring show up on a smoke test?