what do these fuel trims mean?

[arcsnsparks]

hello all

 

i was hoping someone could shed some light on the data log i will attach to this post. The log is from my 2008 outback xt with 5EAT

 

mods are: cobb down pipe, cob intake with a custom tune

 

i'm concerned about the AF learning 1 B at 18.75 and the AF correction 1 in general

 

the log is my drive home from work...with some idle time at the beginning and end.

 

any insight would be greatly appreciated.

 

https://drive.google.com/file/d/1wS6IZ-fM_5Aevzm0ePcqfzP5dEIh6GiG/view?usp=sharing

Edited July 13, 2019 by arcsnsparks

[awfulwaffle]

Looks to me like you've got some mismatched feedback coming from the sensors which contribute to the closed loop fueling target. That is, the range B learned trim is super rich and the short term trim is slamming full negative to try and maintain target AFR.

 

Curious what the other ranges look like? I run open source, but I recall someone else on here mentioning you don't have an option to pull a learning view with an Accessport. However, from your log it appears that range A is good, and when you start to just get into boost a couple of times your range C learned trim swings well negative.

 

I've seen LTFT and STFT fight like that before before, though not nearly to this extent, with a missing rear O2 sensor. The rear O2 only has a minor impact on the closed loop fueling target, but the front O2 is a major driver. I've had the front O2 fail as well, but I just lost all AF reading there (-20 constant) whereas yours still appears to read something. I've also had an MAF fail and start reading way off everywhere except idle, and both STFT and LTFT slammed full positive, which doesn't match what you're seeing.

 

Either way, my inclination would be to check out MAF and front O2 sensor first. Do you know anyone else locally who has one of these cars who would let you swap MAFs and O2 sensor to troubleshoot? Otherwise, a bunch of Subarus (NAs included) from around the same model year use the same MAF. You should be able to find one at a scrap yard for 20 bucks or so, the Denso P/N is right on the housing so you can be sure you get the right one. Not sure about interchangeability of O2 sensors.

 

May be worth posting up your general location in case someone with a spare is local to you.

Edited July 13, 2019 by awfulwaffle

[arcsnsparks]

Looks to me like you've got some mismatched feedback coming from the sensors which contribute to the closed loop fueling target. That is, the range B learned trim is super rich and the short term trim is slamming full negative to try and maintain target AFR.

 

Curious what the other ranges look like? I run open source, but I recall someone else on here mentioning you don't have an option to pull a learning view with an Accessport. However, from your log it appears that range A is good, and when you start to just get into boost a couple of times your range C learned trim swings well negative.

 

I've seen LTFT and STFT fight like that before before, though not nearly to this extent, with a failing MAF. It read at more or less the factory service manual's spec value at idle just like yours does. Either way, my inclination would be to check out MAF and front O2 sensor first. Do you know anyone else locally who has one of these cars who would let you swap MAFs to troubleshoot? Otherwise, a bunch of Subarus (NAs included) from around the same model year use the same MAF. You should be able to find one at a scrap yard for 20 bucks or so, the Denso P/N is right on the housing so you can be sure you get the right one.

 

thanks for the input

 

ill look into the maf sensor. i have friend at local independent subaru shop that does most of the work the car needs. so i will see if i can swap MAF sensors there.

 

what do you mean by a learning view?

 

 

the car had a revised tune 2 months ago from logs that i sent the tune shop where it was dyno tuned. the LTFT were around 0 at A .....+7 to +9 at B ......-6 to -7 at c and D. i was told "this was as close as he could get it without having it there" after the tuner reviewed the revised log.

 

i live in an area where the nearest tune shop is a 3 hour drive.

 

that negitive value in the STFT B range only seems to be at a cold start

[awfulwaffle]

Check my revised answer - I went back to my notes and realized that failing MAF caused both trims to peg to max, whereas an O2 sensor caused a fight like you're seeing. Not to say that it's the only way a failure can manifest itself, but that's my experience. That is, I'd probably check the O2 first, unless your friend has both on hand in which case I'd check both.

 

In open source land, a learning view allows you to view all learned fuel and timing adjustments across the board, as seen here: http://www.romraider.com/forum/topic2772.html

 

I don't know if there's a way to pull something similar with a Cobb setup, but I've heard on here that you can't with just the base software. If not, you just have to watch the real-time adjustments in a log like you're doing.

 

Based on conventional wisdom, those initial LTFT values you mentioned after the revision are near the limits of the acceptable +/- 5-8% range, but your max adjustment range is also higher than stock in my experience so that figure may not hold. Not sure if that's just a late model OBXT difference, or whether the tuner adjusted the limits (stock limits on my 05 are +/- 15% LTFT adjustment).

 

I'd be curious to know where the trims were at historically, and what the reason for the tune revision was.

Edited July 13, 2019 by awfulwaffle

[arcsnsparks]

thank you for sharing your thoughts on the matter. it is appreciated

 

i have had this car for about 8 months

 

LTFT after the dyno tune were always positive. typically +3 +7 . i only saw LTFT rise to 10 a few times when i was on a long drive over a mountain pass.

 

the reason for the revision was due to an modification i made to the intake. the tune shop that put on the cobb intake had the MAF sensor facing down and it was also pressed into the air box hard enough to cut into the plastic housing of the MAF itself. also the cone filter was pushed against the inside of the fender well and distorting and wrinkling the filter. so i trimmed the MAF housing half an inch to give the air filter some room and re worked the mounting bracket for the filter and MAF housing so the MAF sensor was accessible without taking the intake apart. they also neglected to put a gasket between the down pipe and mid pipe. i had to pull the old one off of the stock down pipe and put it on for them and they had to ship the plastic engine cover to me because the did not reinstall that either.

 

that change i the intake plumbing had my LTFT A rising to +18. the revised tune was great for about 6 weeks

 

no the access port does not do a learning view.

[DrD123]

If your fuel trims are positive, then the based on the oxygen sensor, the car is running lean and having to add fuel - what does the STFT do as you increase engine rpm (try to stay out of boost) - if it goes down, that's a strong indicator of a leak. Another issue could be modifying the MAF sensor region - the MAF sensor needs the region it is located in to have predictable flow across the entire diameter (and the same as the factory system) - it takes a reading, then the car calculates how much air is passing through given what it knows about the air intake and what the MAF reading is - if you change the flow pattern, then the car will have trouble and miscalculate how much airflow is actually there. Bottom line - if the car is always adding fuel (large positive LTFT+STFT) then there is more air getting in then the car thinks is getting in, so it adds fuel until the af ratio is where it wants it to be (at the O2 sensor)