O2 code keeps coming back

There's over a dozen oxygen sensor fault codes, and they mean very different things. Erasing a fault code doesn't make the defect go away. If you can post the exact fault code number, I'll find the definition for you and we'll figure out where to start the diagnosis.

Do you have a 3.5L engine?

P2377.

After researching this fault code, I know even less than before. This is very frustrating because the code means different things on different brands and models of cars, and there are no definite diagnostic steps other than to check a lot of stuff. One article mentions a loose or leaking gas cap, an exhaust system leak ahead of the front oxygen sensor, "cylinder 4 contribution/balance", and "ignition coil 'A' primary or secondary fault", all for the same code. None of those are helpful or give us an idea of where to start looking. They do mention that this is not a serious fault code, but it is very common.

I'm going to get one of our other experts involved to see if they have run into this and have a better idea of how to diagnose this code.

Could you post the freeze frame data for the code please? You say it comes right back, after how long and is it consistent or repeatable? IE you clear the code, start the engine code comes back as soon as it goes closed loop, or code returns as soon as you rev the throttle once the engine is warm? Anything like that. Talking to a couple Honda/Acura folks and one of them mentioned that he saw that code when there was an issue with a crank sensor that was not reading correctly and was causing an ignition coil problem because it would trigger the coil twice. Beyond that is the same problem Doc ran into, no good code information. Even in the service manual for that vehicle there isn't any good guidance.

I see. I will get you the freeze data and go from there. But wouldn't a crank sensor if not reading correctly affect it when starting it when warm? Just thinking out loudly.

Cleared it. Started the car. Throttle doesn't affect it. Waited 15 seconds and cane back, did it again and again it cane back in 15 seconds. Attached

More.

It could be not always. In the case he was talking about it wasn't a constant failure but sending odd data. So instead of a pattern of data like 1-2-3-4-1-2-3-4 it would send 1-2-3-3-1-2-3-4 instead. However it's hard to know for sure when there is no real data on that code.
Do you perhaps have a second scan tool that can read just the code? Just want to verify the number. I don't want to give you info on a code only to discover it was something like a coil failure.

Yes, i have other scanners, just the fault code or freeze frame data?

Attached.

Okay, those numbers let me find a bit more. Acura shows the 155-14 code as almost identical to a P0155 code, except it is an issue with the Os sensor being shorted to a higher voltage. I would verify the ground for the sensor as a bad ground could cause the signal to be elevated. If the ground is good then monitor the sensors output to see if it's correct and changing the way it should. My bet is that it is showing a voltage higher than it should and that is setting the code. Check the green wire to the sensor for a problem.

Thanks. How to check the output? I will ground is good and green has power, these tested while car running? Measure output via scanner? When cold? Operating temperature?

Being an AFR sensor and not a standard O2 sensor they can be a pain. The scan tool should read the lambda, amps and voltage from the sensor. Looking at that second image showing those numbers you can see there is no ma draw on the bank 1 sensor. That makes me think it has either a bad ground or a bad connection. The Lambda of 1.00 also makes me wonder as that isn't something that normally happens unless the mix is perfect. Lambda at perfect stoichiometric is ­displayed as lambda 1.0. Numerical values above lambda 1.0 indicate a lean ratio, while ­values below lambda 1.0 indicate a rich ratio. If you look at bank 2 it's showing 1.10 or slightly rich. A quick test is to run the engine until a bit, they normally start working at 10-20 seconds of run time, now stomp the gas pedal to create a rich mix and watch that lambda number. It should change to a rich mix and then settle back fast if the sensor is working.
Much more information on them and what to look for.
https://www.youtube.com/watch?v=crGVIneq1wo

I thought the sensor after the catalytic is a normal sensor, or is it the other way around?

The code you have is for B1 S1 which is the upstream AFR sensor. Upstream should be a 7 pin connector while the downstream is a 4 pin. Both of the upstream units are the same so you could swap the sensors around and see if the code changes. That would narrow it down a bit.

Good idea, thanks.

No problem, I do not like spending money if I don't need to.