O2 Sensor Reading Interpretation

[kstalker]

Hi Folks, I'm trying to interpret some oxygen sensor reading for my '99 TJ Sport 4.0 auto. With the engine warmed up or cold, the bank 1 sensor 2 reads close to 1.0 at idle and is consistently much higher than sensor 1, The attached images are the readings at idle and when revving. Is it possible that sensor 1 is not reading right, causing a rich condition which is showing on the sensor 2 reading? Or could sensor 2 be faulty, causing some other issue?

The TJ runs OK, but feel a little sluggish, and has a little more vibration from the engine than I'd expect. 12 MPG currently.

 

[macleanflood]

My unlearned opinion is that they're trending below .9v and ok.

Yes there is a much more complicated answer.

-Mac

 

[GoldenGorilla]

That doesn't look right. The downstream usually isn't moving as fast as BnS1, but isn't pegged.

Is the jeep new to you?

Did you install this rear O2 sensor? Do you know which brand it is?

Did something change about the way it was running?

It's not unusual right on initial start/warmup to run rich. The PCM does that to get the cat efficiency going faster. The engine runs on open loop for a minute.

 

[hear]

Geez we need to train a model on how do diagnose O2 sensor readings. It's all just jagged graphs and all we really seem to know is that it has to stay within some range. If only we had something like a computer that could be trained to figure it out.

 

[kstalker]

That doesn't look right. The downstream usually isn't moving as fast as BnS1, but isn't pegged.

Is the jeep new to you?

Did you install this rear O2 sensor? Do you know which brand it is?

Did something change about the way it was running?

It's not unusual right on initial start/warmup to run rich. The PCM does that to get the cat efficiency going faster. The engine runs on open loop for a minute.

New to me, had it for a few months. Bought it with a cracked exhaust manifold, so it’s running better now, but still sluggish. These readings were taken in closed loop and operating temp. Agreed that being somewhat pegged close to 1 seems off, would expect a little movement and not always be above the upstream reading. No clue about the brand or age of the downstream sensor

 

[lBasket]

Geez we need to train a model on how do diagnose O2 sensor readings. It's all just jagged graphs and all we really seem to know is that it has to stay within some range. If only we had something like a computer that could be trained to figure it out.

Time to crowd source a training dataset

 

[hear]

Time to crowd source a training dataset

need at least one of us that runs perfectly clean, lol.

 

[lBasket]

need at least one of us that runs perfectly clean, lol.

Maybe just a classification model of good or not? Anyone without a CEL for O2 sensor could help, might have a limited set of people that have known bad O2s but maybe it can know if not group A then group B? I have not worked much with classification past basics

 

[hear]

Maybe just a classification model of good or not? Anyone without a CEL for O2 sensor could help, might have a limited set of people that have known bad O2s but maybe it can know if not group A then group B? I have not worked much with classification past basics

Same. I just sell all this tech, I'm a real laggard on actually using it. (although I do have something cool for our How To indexing coming out soon).

Collecting the learning set would be the hard part, how do we get a bunch of old farts to collect data and submit it. And the 02 data by itself isn't probbly good enough. Probably need rpm, timing advance, throttle position, engine temp, ambient temp, & who knows what else since they all interplay.

 

[macleanflood]

Well I think the first thing you want is the logic tree the PCM uses to throw O2 related codes.

-Mac