2016 Subaru Liberty (Legacy) 3.6R (EZ36D) P0174 - System Too Lean (Bank 2)

Hi all,

I believe I've got an issue with my A/F sensor on bank 2, I wanted to throw some info out there to get some thoughts.

Also a note, I think something like this might make for a good tutorial video. While there are a lot of videos out there with P0171 codes for four cylinder engines, or P0171 and P0174 codes being reported at the same time, I couldn't find many that specifically focused on just P0174 and the diagnostic process.

Sensor layout: An A/F sensor in each of the headers, an O2 sensor in/after each of  the catalytic converters.

My understanding of the sensors (Please correct if I'm wrong, then probably ignore the rest of what I've written HAHA):

- The A/F sensors are trying to reach the lambda value of 1.000.
- This value will almost never actually be met, and the value will flap between just below and just above (0.98x-1.02x) the lambda value.
- The O2 sensors will read between 0v and 1v, with an ideal value around 0.5v.
- The factory service manual provides a reference voltage of 0.750v (From memory).

Info/Questions:

No other error code was ever reported, so while it was suggested, I was less inclined to believe the issue related to the supply/flow of air or fuel. To test though, I did clean my MAF, which was reading around 4.5-4.9g/s with the A/C off, and tested fuel pressure, which was around 340 KPA and didn't dip while racing the engine. I also sprayed starter fluid in various locations between the air filter box and the head's air intake manifold and monitored for changes to idle or sensor readings and found nothing.

So initially I was focused on the O2 sensors. The O2 sensor for bank 1 was reading around 0.750v, though the bank 2 sensor was reading around 0.980v. Realising this meant that the exhaust was very rich, coupled with a short and long term fuel trim of around 35% each and a heavy fuel smell, that all lined up.

While researching the A/F sensors further, I tested disabling injectors one by one on each side of the engine to see how the computer responded. I found that if I disabled either cylinders 1, 3, or 5 (One by one), the bank 1 O2 sensor would dip from around 0.750v to <0.1v. When disabling either cylinders 2, 4, or 6 (One by one) however, I noticed that the bank 2 O2 sensor value dipped from around 0.98xv to between 0.800v and 0.850v, and began to climb.

After this, I found information suggesting that the A/F sensor will report >1 if the exhaust is lean, and <1 if the exhaust is rich. This info also suggested that this value should flap just above and below 1 constantly, because it can never really achieve a stable 1 value. While checking the A/F sensor on bank 1, I found that while not appearing to constantly change between >1 and <1, it was flapping up and down almost all of the time. When looking at the bank 2 sensor however, I found that while sometimes the value would fluctuate, and most of the time that fluctuation would be when increasing rev's from idle (700 RPM) to between 2000-3000 RPM then dropping back to idle, the value was generally stable >1. Most of the time it was reading either 1.01 or 1.02 from memory.

Based on all of this, I believe the issue to be the A/F sensor on bank 2 and will probably swap it with bank 1 before purchasing a replacement, but I wanted to get thoughts from the community here on what else you think it might/could be, what other data I should be looking at, etc.

Cheers in advance!

I would deal with the p0174 code the same way that I deal with the p0171.

I think swapping the AF ratio sensor is a valid test. If the problem moves, then you know it's the sensor.
I would caution that just because only one bank set a lean code doesn't mean both banks aren't running lean.
I would definitely have a look at the fuel trim for both banks, it's not uncommon for one bank to set a lean or rich code before the other.

Noah wrote:

I would caution that just because only one bank set a lean code doesn't mean both banks aren't running lean.
I would definitely have a look at the fuel trim for both banks, it's not uncommon for one bank to set a lean or rich code before the other.

Hi Noah,

The reference guide in the factory service manual for the car lists the bank 1 O2 sensor as reading 0.7v, and the bank 2 O2 sensor as reading 0.795v, so it does appear that while they're certainly higher than what is typically referenced in documentation for O2 sensors, the values for bank 1 are closer to the reference value.

https://lh3.googleusercontent.com/pw/AP1GczOfGPgU-3Bi3YANDU9k9xipIIFZXrNPmzcfnhBtgnWYgR4yfnhl7PvPpBHpT1AN-ta4oEFTgA-7dKSc5S1qPi3xVhbHLavGY8emO_AgTWSupxWO8S90blFK21qOXiZq_vnmIP5T0dQj12S2rYvI4RZKiQ=w1209-h84-s-no-gm

Here's a sample of some of the short term fuel trim for banks 1 and 2:

https://lh3.googleusercontent.com/pw/AP1GczM3_fmugT0pGMF-jf2YKHhEsmYyf6Kjs7fP8Ji64W2b6gU9-XSXYQHvWpsWVQYNOBHDAjq0-M8Qe1OW5myKYaRodrGqOi83NBwLW1OhvGxzA-pLlzPq=s0-d-ip

And long term fuel trim data for banks 1 and 2:

https://lh3.googleusercontent.com/pw/AP1GczNoVvzXwzwDmLL9-SQLP45VGCdIELdRJlAR-RAQCbq0Ma7sr5zkloBH99mM1DLmNjTwgUWMLqAWymPuqeH_ykGejG0i9OF59A6I7R8DD9D9M7f1eZj0=s0-d-ip

Along with some O2 sensor data for banks 1 and 2:

https://lh3.googleusercontent.com/pw/AP1GczM5YMObGv6GXB3XuK-D-5ISiAdWLCUuFuIUbKbVjIZea1uvX55hKpKrfDkhx-MuB-wkcs8Fse6qm-LBmiGH4y57mY5nNG_02FnyO7Mbpn4r45WoGIsg=s0-d-ip

And finally A/F sensor data for banks 1 and 2:

https://lh3.googleusercontent.com/pw/AP1GczOpELI2y_8BAaT4R2cgGTnJ8EPCD4-AIhtExP1XQPPMc1Koy1rdO4Ed_8l2QKjsWc9X4sVuH7K3bWIN_oG7ppTIE4OLe_sPqdZPcyzUAtN3jWpLM_IS=s0-d-ip

I hope this might provide more insight, if there's other data that might help please let me know, but from what I can see with the data above, bank 1's O2 sensor is rising and falling, as is the A/F data, where the data for bank 2 is mostly consistent for both the O2 sensor and the A/F sensor.

There were some cases when racing the engine when the A/F sensor would move a little, as did the O2 sensor, but they look much less like bank 1. To my untrained eyes, bank 1 might look a bit high but might be within spec, where bank 2 only appears to be maxed out.

Other than possibly a leaky injector, the only other thing that would make sense to me is the A/F sensor, but again, that's why I'm here I'm not a mechanic, just a guy who likes to fix his stuff, and with the internet as my instruction manual. Advice from people who know how to read the data better than me is always appreciated

I've attached the photos, I'm not sure why they didn't show up in the post, they were in the preview.

I can't exactly tell if all of those graphs are from the same point in time, but I would agree with your assessment.
If the short term fuel trim is +35% and the downstream O2 is .900v, then it looks like the AFR on that bank is misreporting.