Engine stall when going into neutral, don't know what to anymore.

[Patarrick]

For the past 2 years I've been fighting this stall issue since I've had the car and I have already parts grenaded the car with no luck.

 

Issue: When slowing down and coming to a stop (in-gear) I'll depress the clutch to go into neutral and the engine will stall out. The stalling is not consistent and does not happen in any regular pattern. The car will start right back up with no issues and drive perfectly fine. No check engine light has ever come on related to this issue. At a stop, it idles just fine with no noticeable issues.

 

The car has 160k miles on it and the engine has 3k.

I've replaced the stock engine with a new built IAG motor and the issue has followed from the old engine to the new one.

 

What I've done to the car to try to solve the issue:

New OEM fuel injectors

New Deatschwerks fuel pump

New OEM Coil Packs (was needed anyways)

New OEM STI Spark Plugs (was needed anyways)

New ground straps

Cleaned MAF sensor

Removed and Cleaned Throttle Body

Retune car at dyno shop (Boost Controlled Performance)

Taken it to the dealership multiple times with no luck

 

There are no known vacuum leaks and the car pulls vacuum just fine.

I see lots of posts suggesting IAC valve but I they've all been about the older WRX motor and none about the legacy motor.

 

I don't know what to do. I'm sick and tired of this issue and nothing I do seems to make it go away.

[derp]

lightweight flywheel or crank pulley?

[Patarrick]

lightweight flywheel or crank pulley?

 

ACT StreetLite Flywheel ~ 14.4lbs

I swapped the clutch about a couple weeks into owning the car so I have no experience or recollection with the older clutch. New clutch is Exedy fjk1001.

 

Stock Crank Pulley

[derp]

My car randomly does it too. To my largely layman mind, it seems like the RPS drop too suddenly for the ECU to keep up sometimes and it just sags too low and turns off. I have a pro-lite flywheel with a single-mass clutch. I wonder if they correlate.

[chhrisd]

Have you tested for a failing alternator/battery? Possibly a defunct ground somewhere?

[Patarrick]

My car randomly does it too. To my largely layman mind, it seems like the RPS drop too suddenly for the ECU to keep up sometimes and it just sags too low and turns off. I have a pro-lite flywheel with a single-mass clutch. I wonder if they correlate.

 

I thought maybe so too but I've coasted down 1000 rpm in-gear then clutched in to neutral and still had the engine stall out even though the rpms were at idle speed.

 

Have you tested for a failing alternator/battery? Possibly a defunct ground somewhere?

 

Battery and alternator both work just fine and neither shows signs of failure. This issue has also persisted for over 2 years now, I'm sure an alternator would've given out completely by then. Also, no torn or missing ground straps, I replaced all of them some few months ago trying to solve this issue.

[serx7]

My car has exhibited this too. In my case a couple of things seem to contribute to it:

 

1) fuel trims 'too far' out of whack. Before I spent a fair amt of time getting closed loop MAF scaling dialed in, the idle and cruise fuel trims (AF Learning 1A and B) were somewhat out of spec. 1A would hover b/n +6 and +8 in warm temps, while 1B would settle out around -3 to -5. When I'd slow down in-gear until ~1000-1200 rpm and then disengage the clutch, it would frequently shake a bit and would sometimes stall. I suspected that the trim swing as it came out of the 1B trim and into the 1A trim after disengaging the clutch, was contributing to the behavior. After a fair bit of CL MAF scaling, 1A goes b/n -1.5 and +1 depending on ambient, while 1B settles out to around -1.6. Getting those trims closer to 0 and closer to each other, definitely seems to have helped the idle/close to idle behavior (cool or warmed up engine).

 

However...

 

2) I noticed that after an ECU reset/reflash and w/ the engine warmed up, the shaking/stalling tendency returns, until things get to the point that the 1A and 1B trims get settled again. I'm not sure that it's just the fuel trims, since they settle out pretty close to 0 and pretty close to each other, but once things have settled after a short local drive, the idle returns to being very stable and the shaking/stalling tendency seems to completely disappear. This has always made me wonder if the ECU learns more than just the fuel trims after a reset/reflash. If not, then perhaps my car is just particularly sensitive to the low airflow fuel trims.

[Max Capacity]

serx7 How close are you to www.AZPinstalls.com ? they are in northern NJ a great place to take your Subby to have it fixed.

[solidxsnake]

My car has exhibited this too. In my case a couple of things seem to contribute to it:

 

1) fuel trims 'too far' out of whack. Before I spent a fair amt of time getting closed loop MAF scaling dialed in, the idle and cruise fuel trims (AF Learning 1A and B) were somewhat out of spec. 1A would hover b/n +6 and +8 in warm temps, while 1B would settle out around -3 to -5. When I'd slow down in-gear until ~1000-1200 rpm and then disengage the clutch, it would frequently shake a bit and would sometimes stall. I suspected that the trim swing as it came out of the 1B trim and into the 1A trim after disengaging the clutch, was contributing to the behavior. After a fair bit of CL MAF scaling, 1A goes b/n -1.5 and +1 depending on ambient, while 1B settles out to around -1.6. Getting those trims closer to 0 and closer to each other, definitely seems to have helped the idle/close to idle behavior (cool or warmed up engine).

 

However...

 

2) I noticed that after an ECU reset/reflash and w/ the engine warmed up, the shaking/stalling tendency returns, until things get to the point that the 1A and 1B trims get settled again. I'm not sure that it's just the fuel trims, since they settle out pretty close to 0 and pretty close to each other, but once things have settled after a short local drive, the idle returns to being very stable and the shaking/stalling tendency seems to completely disappear. This has always made me wonder if the ECU learns more than just the fuel trims after a reset/reflash. If not, then perhaps my car is just particularly sensitive to the low airflow fuel trims.

 

My car exhibits exactly the same behavior, on all accounts.

[serx7]

serx7 How close are you to www.AZPinstalls.com ? they are in northern NJ a great place to take your Subby to have it fixed.

 

I've done business w/ Mike, Rey et al at AZP in the past w/ my Saabaru (timing belt and brake work). Good crew there. AZP actually put me in touch w/ the PO of the modded/VF52'd OBXT I drive now, 3+ yrs ago.

 

As far as I'm concerned the OBXT is fine, the shake/stall thing only comes up right after an ECU reset or a reflash (I self tune so reflashes are somewhat frequent). Once the car is allowed to warm idle for 20-30 or so sec (usually after the 1st stoplight after the coolant temp is up), the behavior disappears and won't reappear unless I reset or reflash again. It smoke tests fine, compression on the built motor is 125 across the board w/ leak down at 5% across the board (same results over the past 3 yrs), and it runs/pulls quite well.

[BrandonspecB]

After a reflash the ECU has to learn how to idle. Learning of any sort doesn't happen until coolant temp is 169 degrees. To avoid the rocky return to idle just let the engine warm up to 169 and let it idle for 30-60 seconds after that. Give the throttle a quick blip to see how it drops back to idle. Should work after that. If not, let it idle for a bit more.

[serx7]

Yeah I noticed AF Learning 1A starts to move once coolant temp gets into the 160F+ range, although I didn't know whether there was idle-specific learning that needed to occur (seemed like it based on observation though).

 

I don't like idling it long enough to learn idle/1A trim from a cold start (after a reflash), so I usually idle it 2-3 minutes (built motor) or until the dash coolant needle starts to move upward and then drive gently and just put up w/ it.

 

Lately what I try to do is make any tuning/revision changes before the motor/coolant temps have dropped too much, then flash the revision, start it up and let it idle for a minute or 2. When I do things that way, I avoid the idle nuisance after the next cold start, albeit at the cost of an additional motor start. Since I work remote I can sometimes do another quick round of datalogging before calling it quits for the morning/afternoon, so sometimes the extra start has additional value.

[serx7]

I wonder if my Target Throttle (Idle Airflow Target) might need some adjustment. I wouldn't consider my build 'big', and it's definitely not rotated, but the symptom seems to be exactly the same until the ECU learns around the idle condition.

 

http://legacygt.com/forums/showpost.php?p=5472865&postcount=149

 

Per Mike @ Tuning Alliance: "Target throttle angles (idle airflow target) - Some engines require increased throttle opening to return to idle smoothly. They often will dip down to 300-500 rpms when you let of the throttle, then come back to target idle speed. They will often learn to give additional angle over time, but some cars will even stall due to this. It generally is only a problem on big rotated builds with after market intake manifolds."

[solidxsnake]

I wonder if my Target Throttle (Idle Airflow Target) might need some adjustment. I wouldn't consider my build 'big', and it's definitely not rotated, but the symptom seems to be exactly the same until the ECU learns around the idle condition.

 

http://legacygt.com/forums/showpost.php?p=5472865&postcount=149

 

Per Mike @ Tuning Alliance: "Target throttle angles (idle airflow target) - Some engines require increased throttle opening to return to idle smoothly. They often will dip down to 300-500 rpms when you let of the throttle, then come back to target idle speed. They will often learn to give additional angle over time, but some cars will even stall due to this. It generally is only a problem on big rotated builds with after market intake manifolds."

 

I didn't have any luck with that table when I still had an AP (in fact, Mike dynotuned my car and did play with that table a bit) changing the unlearned idle response... after an ECU reset it still idled like crap until it was re-learned.

[serx7]

I didn't have any luck with that table when I still had an AP (in fact, Mike dynotuned my car and did play with that table a bit) changing the unlearned idle response... after an ECU reset it still idled like crap until it was re-learned.

 

Bummer. Maybe during lunch today I'll add Target Throttle to my AP3 log list and do some operating temp datalogging at/around the neighborhood stop signs (where I always encounter the behavior when heading out after a reflash). I'd be curious to compare the data as-is (learned) vs right after a reset but w/ the car still at operating temp w/o giving it a chance to learn anything. Maybe something useful would show up.

 

More likely, I'll just eat lunch at my desk and read other threads here on forums and deal w/ the reset/reflash idle quirk. Been living w/ it for the past 3+ yrs

[Tuning Alliance]

Its fixable, i have seen this on basic modified cars as well. Sometimes your need to edit the primary throttle tables and idle airflow targets.

 

Shoot us a PM and we can get you started on a etune.

 

-Mike/Brian

[M4rked1]

Not sure if it is related, but I had very similar issues of randomly stalling right before my OCV went bad. When it finally died the car could hardly idle in any situation, but just as it was starting to go it would behave similarly to what you describe.

[Patarrick]

serx7 highjacking my thread, lol. It's okay though, I've been absent this past week.

 

I'm gonna try getting a log running on my car and see if I can capture a stall and see what's going on with it. Is there anything in particular that I should be logging?

[Tuning Alliance]

serx7 highjacking my thread, lol. It's okay though, I've been absent this past week.

 

I'm gonna try getting a log running on my car and see if I can capture a stall and see what's going on with it. Is there anything in particular that I should be logging?

 

A/F Correction #1

 

A/F Learning #1

 

AFR Ratio Sensor #1

 

Boost (Manifold Rel. Pressure)

 

Calculated Load

 

Engine Speed

 

Feedback Knock Correction

 

Fine Knock Learning

 

Ignition Timing

 

Intake Temperature Pre-Turbo

 

Inj. Duty Cycle

 

Mass Airflow

 

Mass Airflow Voltage

 

Primary Ignition

 

Throttle Position

Waste gate Duty

[Patarrick]

Annoyingly, when I started logging, the car did not want to mishave and acted like a perfect gentleman >:|

 

But I finally got a stall after days of logging.

I've uploaded the log onto google drive, link below. The stall is at the very bottom of the spreadsheet, at around row 5200.

 

https://drive.google.com/open?id=0B9esEHgnhEU5Q1FRdUhQclRNc2s

[MaStaMooN]

Mine has never stalled but dips pretty low since going with a lighter weight flywheel.

 

Could it be something like a flakey cam or crank position sensor? Or a fuel regulator.. Weird.

 

Sent from my Nexus 6 using Tapatalk