af correction woes.. wits end

I'm having an issue with AF correction (STFT). I posted a video to youtube as it's the easiest way to show what's going on. I know video is shaky, but just look at the top left gauge.

 

[ame=http://www.youtube.com/watch?v=S66JWshNCIk]Af correction issue - YouTube[/ame]

 

In 5 years of monitoring and adjusting af correction on this car, I've never had this happen. It's making it nearly impossible for me to get anything dialed in.

 

It's rapidly moving between +4 or 5 and -1 or 2 while at steady throttle. Typically at steady throttle it will be somewhat steady.

 

I believe it is a minor boost or vacuum leak, but I haven't been able to find anything wrong. I've replaced the BPV with aftermarket. I've replace the TMIC to Turbo gasket with the mr gasket. Checked and zip tied all vacuum lines. Removed, clean and reseated the maf. I sprayed all the connections with Carb cleaner too and noted no change.

 

The few things I'm considering trying are:

 

New front o2 sensor.

New Maf sensor.

New PCV valve.

Different TMIC gasket.

Different TMIC.

 

I'm hoping someone has seen this issue and resolved it?

Relavent mod list:

 

KSTech 73mm

BNR 16g

GS EBCS

IPR TMIC

GFB BOV (also tried 2 stock units)

Catless Downpipe

DW 750s

DW65c

E85

I do.. They're within reason of each other. They've never been dead on even after replacing the WB02 sensor with new..

 

EDIT:

 

3rd gauge over from left on top row is OEM O2, and last gauge on bottom is WB02.

Just throwing this out there since it seems you have most everything else out there. Any pre front o2 sensor exhaust leaks? At low loads the exhaust pulses create a vacuum and can pull air into the exhaust causing the o2 to read lean. This would cause a positive correction in closed loop.

OP, off-topic:

what set up are you using in terms of tablet PC and software to view the real time data? I'd love to do this!

OP, off-topic:

what set up are you using in terms of tablet PC and software to view the real time data? I'd love to do this!

 

It's just a laptop in trunk hooked up (video only) to the OEM nav screen. I use a Logitech Dinovo mini to control it. Tactrix/WBo2 are hooked up to it and I mainly only use it for data logging/data display. Occasionally on long trips it does double duty as a media player.

Something else I thought of. What does maf voltage and airflow do when af correction is acting funny? Do they fluctuate at a steady engine speed and throttle? Could indicate a faulty maf sensor.

Thanks for the suggestions BKatter. I will check around for any potential leaks pre o2. Maf g/s seems to hold fairly steady. It's actually one of the parameters in the video (bottom left gauge). I will pay more attention to it next time I drive the car on the highway.

 

I posted this to a local subaru forum as well, and a local tuning shop that most people here use thinks it's my GFB valve leaking. He says these valves are junk, which was surprising to me as it's touted as and seems to be a fairly quality piece. He has tuned thousands of cars though so I have to at least consider his opinion. I might add a TurboSmart Kompact to my list of parts to throw at this problem.

Did you ever figure this out? Mine seems to do it too, but never thought it was a high issue as long as the corrections aren't extreme.

 

Are you sure you're not thinking of a/f learning which should remain constant under specified airflow ranges?

No.. Still haven't gotten anywhere with this. I'm certain the behavior is not normal. If you're holding at says 2400 rpm and .7 load, you should get a reasonably steady af correction reading that would allow you to make adjustments rather than a wild swing between -1 or -2 to +5 or +6..

 

AF Learning ranges are all staying near 0 though because it's not steady enough for the ECU to make changes to LTFT.. If you look at the top 2 boxes on the left in the video those are ST and LT fuel trims.

 

I've got an AVO TMIC, AVO TB hose and new Tbolt clamps on the way. I'm hoping this solves the issue but I'm some what doubtful.

FYI, AVO TMIC lower bolt hole for the BPV/BOV goes all the way through into the end tank, allowing a small leak. Stupid design in my opinion. Seal it up by putting on RTV and be careful of stripping it.

Thanks INVAR. I saw this same issue with the IPR TMIC and ended up JB welding it from the inside and using a smaller bolt that wouldn't go through. Not sure if I will address the AVO in the same way or use the RTV as suggested.

 

I did pull off the OEM TB hose today and noticed that it was fairly loose on the TB and it was torn fairly bad. I don't think the tear went all the way through yet, but it doesn't look as if it would have held up much longer.

FYI, AVO TMIC lower bolt hole for the BPV/BOV goes all the way through into the end tank, allowing a small leak. Stupid design in my opinion. Seal it up by putting on RTV and be careful of stripping it.

 

Yikes, didn't know this! I installed helicoils (thread inserts) right away to avoid stripped threads in the soft aluminum but didn't remember the hole going all the way through.

 

TB hose: I have an AVO aftermarket TB hose and really like it...seals better than stock and is a little longer (stock was always a little short in my opinion).

I swapped out the OEM TB Hose with the AVO (what a pain!) and added 2 T-clamps in place of the OEM clamps. I also swapped out the IPR TMIC for the AVO TMIC.

 

I made a 3rd gear log on the same stretch of road as always and it was within a few degrees temperature wise. No tuning changes were made.

 

Ref HP and TQ were 333/364. New HP and TQ were 345/370. 60-80 time went from 2.28 to 2.14. With the IPR I've done dozens of pulls on the same stretch since switching to E85 and they always came in around 330/360 so either the TB Hose or the AVO in place of the IPR show some minor improvement. Whether it's a result of better hardware, or the fixing of a minor leak I'm unsure.

 

I didn't get a chance to do a lot of highway driving, but the wild AF fluctuations seem to be reduced. It's been a bit intermittent so I can't say for sure without a few long rides.

When I swapped the TMIC, I put a new OEM BPV on for testing. All has been well for a few days but it started not holding the power after a few wot logs. I switched back to my GFB and the problem appears to be back. I'm not sure if it's the valve itself or the gasket that they provide. I believe it's the gasket but not sure how to improve it.

 

The AVO BPV opening is oddly shaped compared to the GFB gasket so I guess it would make sense that the gasket is the issue. I'm either going to try using some RTV or look into a different BPV.

What boost are you running? I think my aftermarket BOV may have issues and swapping to stock just to check.

Right around 21 psi peak tapering off to 17 at redline.. I really want to run the OEM BPV but it's not going to hold those levels.

http://www.infamousperformance.net/servlet/the-655/Turbosmart-Subaru-WRX-%26/Detail

 

Thanks. I've actually had that valve before. I decided I wanted something that had easily adjustable spring tension. I'd thought about getting another, but then I saw Underdogs thread that shows the Sti version of this valve leaking fairly badly at idle.

 

http://legacygt.com/forums/showthread.php/fs-ma-turbosmart-kompact-recirc-bpv-0203-1215-214636.html