Weird FKL/AF Learning/AFR Conditions - Exhaust Leak?

[amm203]

I've been tracking an issue that is getting progressively worse. I was originally thinking a bad batch of gas but, given the recent values I data-logged, something else seems wrong. Couple data-logs are attached but essentially I'll get -.35 FKL, +10% AF Learning and -5 to -14% AF Correction all at the same time. This doesn't seem right and was not the case a few weeks ago. Reading around I'm seeing anything from bad FPR to exhaust leaks and everything in between. Any help is appreciated!

 

Edit: Forgot to mention it appears the engine is struggling to maintain a good AFR (routinely well below 14.6).

datalog4.csv

datalog3.csv

Edited October 31, 2018 by amm203

[xt2005bonbon]

Have not looked at your logs but have you checked for vac leaks? MAF clean? MAF wiring harness not acting up?

[amm203]

I haven't checked yet but it idles at -9-10 psi FWIW. I'll try the old "plug the intake with a can and blow in the BPV vac line" trick this weekend. Per my tuner's advice, I'm also going to run the tank really low, refill with Shell or something else good, reset the ECU and log. I'll report back after that!

[xt2005bonbon]

mmmh. Why do you want to reset the ECU? I personally would not. Is your IAM still 1?

[Cutlassjim]

What's the injector duty cycle look like? Are you logging fuel pressure? Dying pump or clogged fuel filter?

[amm203]

mmmh. Why do you want to reset the ECU? I personally would not. Is your IAM still 1?

 

DAM is at 1 throughout. Figured doing so would help but if not then alright. Better off just logging only?

 

What's the injector duty cycle look like? Are you logging fuel pressure? Dying pump or clogged fuel filter?

 

Duty cycle looks normal as far as I can tell. Running ID 1050X with a pro tune so I'm definitely not stressing them out with my setup. I am not monitoring pressure (don't think I can without an added sensor).

 

Hoping I'm just over-analyzing this one in general. I'll find out after some BP or Shell 93

[xt2005bonbon]

In general, I prefer letting the ECU correct itself. I mean, if it noticed some real knock and learned to pull timing, it is better to leave it this way until the problem is identified/resolved. If the issue is found, then the ECU will slowly add timing back, correct the fuel trims, etc... Actually, the latter is pretty quick. That's why I no longer bother resetting the ECU.

[xt2005bonbon]

Just had a quick look at your logs. As you said, all of your learned fuel trims are positive. So you may well have a vacuum leak somewhere. And yeah, your AF correct does swing quite a bit. In any case, first things first IMO: check for vac leaks. In the meantime, don't push the car too much even if your DAM is 1 at the moment. Edited November 2, 2018 by xt2005bonbon

[relative4]

How old are your MAF sensor and upper O2 sensor?

[amm203]

MAF & upper 02 replaced less than 20K miles ago. Anything I should look for while monitoring the MAF voltage and g/s to see if something's wrong?

[relative4]

Both Denso?

 

Can you post a log including Cylinder 1-4 roughness?

 

Can you post your A/F Learning in the A-D ranges?

[amm203]

-Denso 197-6040 Mass Air Flow Sensor

-Denso 234-9035 Air Fuel Sensor

 

I'll log those today on my way home and post them up right after! Appreciate everyone's help so far.

[amm203]

New logs attached and noted (Cold Start, Warm Drive, Warm Boost - mild 4-5 PSI). I gave the engine a once-over before doing these logs and found 2 odd things. The lowest line at the IAG AOS had a little oil on it. Not sure if that's a big deal. I then noticed that the vacuum line to the MAP sensor was unseated enough where I could see about 1/4" of the plastic nipple. It is zip-tied so I'm surprised it managed to unseat itself that much. I made sure to slide it all the way up tight to the MAP.

 

The cold start is concerning with the roughness. However, it is worth noting I'm running a modified engine with forged internals, one step colder plugs and ID 1050X injectors. Hoping the cold roughness can be chalked up to that because I am getting no roughness whatsoever after about 2 minutes of idling.

 

I think the other logs speak for themselves, though. The car feels like it did right after my dyno tune - power delivery is much much better and the car just feels better overall. I'm curious if that vacuum hose could really be the cause of all of the wonky trims and FKL. To be safe I ran it low and filled up on BP 93. Still going to monitor closely but I'm feeling pretty optimistic.

 

Another note: I recall my builder commenting on having to replace a coil because of misfires after the rebuild. Could another coil be on its way out @ Cylinder 4? Maybe I'll swap it with another coil tomorrow to see if the cold start issue follows it.

Cold Start.csv

Warm Drive.csv

Warm Boost.csv

[Marvick08GT]

The map sensor is the only way for the ECU to monitor pressure.. Keep an eye on it and keep is posted. GL

[relative4]

Lots of worrying stuff going on. Cold misfires, high A/F corrections, running lean at high load... I'd like to see a pull, but I'd start by checking the fuel system. What sort of turbo, fueling, and boost are you running to require the one step colder plugs?

[amm203]

Appreciate the feedback! I thought wide A/F correction swings weren't too uncommon. My A/F learning was -2 to +7 this morning with no FKL and no FBK on a 20 mile drive.

 

VF48, AEM fuel pump with ID 1050X injectors tuned for 20 psi. It was the tuner & engine-builder's decision to run the one step colder plugs.

 

Edit: Forgot to mention I'm running TGV deletes and no air-pump if that means anything. Also, that cold-start was done outside when the temp was 40°. I'm going to do a cold-start log tomorrow morning from my 72° parking garage and see if anything changes.

Edited November 5, 2018 by amm203

[xt2005bonbon]

I think it is a good sign that your AF learning values are getting closer to zero. Now, I think it will take a little longer for trim D to get back to 0% compared to the others.

Regarding your misfire count, keep an eye on it and see if it gets worse during cold idle when the outside temps get even colder. Hopefully, it won't get worse. But a max count of 7 is not toooo bad so long as it does not keep repeating itself. IIRC, a misfire CEL will get thrown if the count gets pass ~32 for a couple of times in a row. Max count reaches 99 btw and start again at 0 if necessary. Hopefully, you'll never get there!

[amm203]

Cold-start (in 72° garage) and WOT pull attached. Still getting roughness on #4. WOT pull was a little short of 20psi but that's my fault in how I revved into 3rd. Car has no problem hitting 20psi. I have the question into the engine builder about the #4 roughness.

Cold-Start Garage.csv

WOT Pull.csv

[xt2005bonbon]

Well, #4 misfire did peak at 8 but it did not repeat itself right after. It just happened once (at least on this log). Then the other cylinders also had some roughness, where #3 had some recurrences throughout the log. So, personally if I see these numbers showing up on all cylinders, then that's somewhat of a good news to me. But yes, #4 had a slightly higher count. But again, only happened once. Just keep an eye on it.

Did you start this cold log right after cranking the engine?

Edited November 6, 2018 by xt2005bonbon

[xt2005bonbon]

Just so you know, when I had a tight exhaust valve on #2, the roughness count at idle would slowly climb to, say, 15 or 20, then went back to zero, then 3 seconds later climbed again, and kept doing that. Then on some days, went passed 30 or even 45, then CEL . Drove like this for quite a while too. But after warm up, she was good. I always thought it was rings cause issues with valve clearance would usually show up during warm idle. Anyway, fyi.

[amm203]

Cranked engine, hit log right away. Car was sitting overnight before doing that. Worth monitoring roughness on warm start?

 

I was kind of thinking the same thing with all cylinders showing a little roughness. Like I said, I'm chalking it up to built engine, colder plugs, etc. but I could be wrong. Considering how much better the engine is running since re-seating the MAP vacuum I'm a lot less worried. (Still can't believe a slightly unseated line would leak and cause so many issues like that but it must have been the culprit).

[xt2005bonbon]

I personally would not log right after cranking. I usually disregard any roughness right then. But that's just me

I would log, perhaps a minute later? When I had my misfire issues, the misfires did not show up right after cranking. They were showing up a bit later when the engine was still warming up and it was idling (at an RPM a bit lower than right after cranking).

 

 

It would not hurt to do a warm idling log just to make sure she is good.

 

 

But yeah, looks like your line that slightly came off was the culprit. Sensitive system for sure!

[amm203]

Good to know because the roughness is pretty much gone after 1-2 minutes. No roughness whatsoever on any cylinder after a warm start-up as well. I'm going to replace these lines with silicone to help avoid this problem in the future.

 

Random thought - always wondered why there was foam around that little white filter on the line coming off the MAP. Realized it must be to limit rattle noises considering its tight location.