Diagnostic help CTS 3.6L

[ChevyHighPerformance]

I have a 2004 Cadillac CTS 3.6L 60K miles. I'm getting P2234 - O2S Signal Circuit Shorted to Heater Circuit Bank 2 Sensor 1. This is a wideband precat sensor on the driver's side.

In the attached log you can see how bank2 wideband has "noise" and bank 1 wideband is clean. (You can see this same noise in the Strims.) The noise on bank 2 actually follows the engine RPM. However, the ECM sees this noise as the heater pulses impacting the O2 sensor and will eventually set the DTC.

Here's what I've done and know:

1) I replaced this O2 sensor with a new bosch sensor. - no change

2) I verified that there are no wiring issues - no change

3) I got a new ECM and programmed - no change

4) I followed the GM DTC procedure. However, I can't run an injector balance test because HPTuners doesn't support this for the CTS.

5) I just started running seafoam in the fuel to see perhaps I have a slightly clogged injector.

6) Last year I logged the Strims (HPT didn't have PIDs for the precat O2s) and this oscillation wasn't there. There was no tuning done since then.

7) I tried telling the ECM to ignore the DTC but it still takes action without setting the CEL. The action is that bank2 runs in open loop. The engine will reset itself only after an ignition reset not while you are driving.

8) Bank 2's lambda oscillates between ~0.97 and 1.03.

9) When I reduce the ignition timing the peak amplitude of the noise is reduced slightly but still there.

10) The car runs fine. No hesitation, no misfires, no knock, no change in fuel economy, etc.

11) I removed the upper intake and replaced with new intake gaskets.

12) I'm running 93 octane fuel.

I'm guesing that the problem could be related to one bank 2 cylinder having a slight issue:

1) Bad/dirty injector
2) Bad plug/weak ignition coil
3) Air leak around injector

Any thoughts would be greatly appreciated.

[ChevyHighPerformance]

Here is another log. At 27:39 is when Strim2 goes to zero and when the DTC would have been set. I also logged the current firing cylinder. Because of the data acquisition rate, the current firing cylinder may not be correct.