[Meech]

My 2016 sti has recently began to randomly (or seemingly randomly to me) decide to run rich (around 11.14 on the air fuel ratio gauge). The car has previously had a small boost leak which I found and fixed. The car ran perfect for two days and while driving on the third day I noticed the air/fuel ratio was alarmingly rich. I pulled over and shut off my car for a few minutes and turned it back on. The air/fuel ratio was still rich at idle, so I decided to drive it about a quarter of a mile to my house and check for boost leaks, but I could not find one. I tightened some of my intercooler charge pipe's t-bolt clamps for ease of mind and turned the car back on. It idled okay but still a bit rich. I decided to take a break and let the car sit for about an hour. I then returned to the car and all of a sudden it idled just fine. I drove the car around the block and everything seemed to be fine. I checked on the car one more time before going to bed and it once again ran perfectly fine.

This morning, I turned on my car and it idled fine. The car drove fine for a few miles. Then the car randomly began to run rich again and bogged down.

I'm not sure what is causing the issue and I need help. My first suspicion was a boost/vacuum leak. However, I smoke/vacuum leak tested the vehicle without findings. The o2 sensors, spark plugs and coil packs were all replaced within the last 3k miles. I did not get new gas right before the problem began.

The only thing I can think of is the MAF sensor needing to be cleaned, which I am going to try on my lunch break. Does anyone have any idea what I could check next? I'm at a loss for ideas.

 

[yamahaSHO]

How did you check for a boost leak? Visual does not cut it.

Outside of a boost leak, running rich at idle would could indicate high fuel pressure, which can be caused by the FPR vacuum reference coming off or having a large leak (or failure of the FPR).

 

[Karlot]

^that.

if you didnt pressure test the system and just pumped some smoke into it for the leak, you may be perfectly fine until there is actually real pressure in the system, THEN the leak would show up.

 

[Meech]

How did you check for a boost leak? Visual does not cut it.

Outside of a boost leak, running rich at idle would could indicate high fuel pressure, which can be caused by the FPR vacuum reference coming off or having a large leak (or failure of the FPR).

I took my air intake off and put a boost leak tester on the turbo inlet. I hooked a compressor up to the tester and pressurized it. I listened for leaks and sprayed soapy water. I didnt hear anything or see any bubbles.

 

[yamahaSHO]

Perfect. I'd start looking at fuel pressure next.

 

[Meech]

Perfect. I'd start looking at fuel pressure next.

What exactly am I looking for in reference to the fuel pressure regulator? Ive visually inspected it and everything looks good. Im not sure if there is a more thorough way to check.

 

[Meech]

Perfect. I'd start looking at fuel pressure next.

Just an update, after making sure the fpr was good I turned on the car and it idled realy rich again. I turned it off and waited 5 minutes and turned the vehicle back on and it idled perfect. Im not sure how the issue seems to be so random.

 

[yamahaSHO]

If you cannot find anything wrong by looking at it, the best way is to use a fuel pressure tester.

The randomness may indicate a lose or rubbed through wire? How is battery voltage?

 

[kopele]

Make sure your MAF sensor is connected well. Give it a tap and see if your AFR goes crazy.

 

[kakarot09]

Are you able to log the car? I'm wondering if it's the sensor itself. If your factory O2 sensor doesn't agree with the wideband, then that can narrow it down a lot.

 

[Meech]

Just an update for anyone following this, I reconnected my factory O2 sensor (I had a wideband sensor connected on that is located on the downpipe0). Having the wideband was causing my car to throw P0138 and P0037 codes. This fixed the codes but did not really help with the air/fuel ratio issue. So I decided to replace the MAF sensor and the issue seems to be resolved.

Thanks you for all the help.

 

[kopele]

You can't just swap aftermarket sensors and expect them to work. You would've saved yourself a lot of trouble if you included all the information in the first post.

 

[Meech]

You can't just swap aftermarket sensors and expect them to work. You would've saved yourself a lot of trouble if you included all the information in the first post.

Interesting, considering switching back to the oem o2 sensor did not fix the problem.

 

[kopele]

Interesting, considering switching back to the oem o2 sensor did not fix the problem.

Sounds like you had two issues with multiple symptoms, which you neglected to mention.
Are you implying that using a wideband o2 instead of the stock narrowband without correct scaling is not problem?
Because your CEL would beg to differ.

 

[Meech]

Sounds like you had two issues with multiple symptoms, which you neglected to mention.
Are you implying that using a wideband o2 instead of the stock narrowband without correct scaling is not problem?
Because your CEL would beg to differ.

It’s almost like a made a previous discussion about the CELs. Didn’t feel it was important to include in the post because I knew it wasn’t causing the specific issue mentioned in this post, and swapping the o2 sensor was the first thing I did after making the post. I didn’t need to be told to do it. It would have been redundant to ask if I should do something I was already going to do. Also I had a tune with correct scaling for the wideband, which is why I was unsure of why the codes were on and made a post about it in the past.

 

[kopele]

It’s almost like a made a previous discussion about the CELs. Didn’t feel it was important to include in the post because I knew it wasn’t causing the specific issue mentioned in this post, and swapping the o2 sensor was the first thing I did after making the post. I didn’t need to be told to do it. It would have been redundant to ask if I should do something I was already going to do. Also I had a tune with correct scaling for the wideband, which is why I was unsure of why the codes were on and made a post about it in the past.

Got it, didn't check your post history, just seemed like you omitted a bunch of stuff.