Resolved: 2005 OBXT Misfiring looking for advice on diagnostics before pulling motor(Xpost from 3rd gen Outback)

[jnstrong]

Issue Resolved

Cross posting here for visibility. 

2005 OBXT 5mt, Looking for advice on what to check next before I commit to pulling motor. 
Current mods: VF52, catless up-pipe, AEM AFR gage, partial tune.

Some Background for context: Bought the car a while back as a project. Clutch was shot, it wasn't idling perfect, and had an oil leak; trailered it home. Car has 125k mi but the engine had just been rebuilt after turbo went, new short block with forged sti crank. 3k mi on the engine, figured it was worth a gamble and got it for cheap. Clutch wasn't as issue, but found the engine installer missed bolts here and there, broke a few minor things, nothing I couldn't fix. Oil leak turned out to be a reman turbo that was leaking internally. I rebuilt an ebay VF52 and installed with intention of leaning to tune the car (Im an ME with professional experience tuning diesel engines). Thanks to all who contributed to the opensource projects like Romraider! After much research and wrenching getting everything buttoned up I started tuning. Idle issue resolved (vacuum leaks), Knock issues resolved with a new knock sensor and wiring. Stopped tuning while establishing the Air-Fuel relationship. Data was showing inconsistency leading me to fuel issues. I assumed the injectors were original. The fuel trims pointed to an over fueling/underperforming injector. I pulled and cleaned the injectors as best I could without a dedicated rig. Reinstalled and resumed tuning, results were good, data was looking good, car felt strong. Fuel trims were within 5 % then the following happened. 

Current issues:
Leaving work car suddenly was hard to start and ran terrible, I had driven around lunch time with a little spirited driving, car was runing well.
Car ran like it was down a cyl, I limped it home and pulled it into the shop. Couldn't get to it for a few days, but Cyl 1 was definitely not firing. Plug was black and wet as expected. Couldn’t do a hot compression test. Cold numbers were terribly low with cyl 2 well out of 20% difference (60-70psi cold). But misfiring cyl wasn't out of line with the average. Quick and dirty leak down of cyl 1 sounded like all air was going past rings, didn't hold pressure except for a few seconds, didn't test other cylinders. During running there has never been noticeable blowby or excessive oil consumption to lead me to a ring land failure. I wanted a warm compression test before pulling engine. Reassembled with og plugs and coils all back in their locations. Used new matched and flow tested 750CC injectors. When fired back up, it started right up, but ran poorly, missing on several cylinders. This time Cyl 4 was dead, intermittent misfires on 1 & 2. I shut it down and haven't messed with it since. Warming the engine on 2.5 cyls seemed like a bad idea.

I think I need to check fuel rail pressure. Sounds like a fuel issue to me, or else all the coil packs need replacing. I got a new fuel pump with the injectors, but haven’t installed yet. The compression numbers were alarming, but inconsistent with the symptoms. Internet opinion was not to trust cold compression on a boosted motor. I hope the rings are not well seated yet. The motor has ~5k on the rebuild at this point. Last 1.5k miles were light driving under only waste gate boost pressure.
What would you do next? 

Somewhat unrelated, has anyone pulled the motor and manual trans as a unit? I need to address a trans leak and potential front diff bearing replacement if I pull the motor.
Thanks in advance!
 

Edited November 22, 2023 by jnstrong
Issue fixed

[subisubisu]

I hope you can get the info you need and track down your troubles, soon.

I did pull the engine and transmission (after a 6MT swap) as a unit from my 2005 OBXT.  I've mated the transmission under the car with the engine in place a couple of times and I much prefer doing it outside the car.  I'd recommend removing the radiator and putting the hood up to its near-vertical position.  I was able to do it with a garden-variety engine hoist.

[jnstrong]

Thanks for the confirmation on the trans. Did you have to pull Ex manifold?

[subisubisu]

10 minutes ago, jnstrong said:

Thanks for the confirmation on the trans. Did you have to pull Ex manifold?

Frankly I didn't remember on the pull (three years ago!) but I checked my phone picture history and apparently I left all the exhaust on it up to the turbo.  I know that when I dropped it back in this summer I had everything on it up to the up pipe.  I didn't want to risk banging up the new turbo on the way in.  So, it can be done that way.  Certainly removing things will make it easier to maneuver in the space available.

[Max Capacity]

If you can get the car high enough, you should be able to pull it as a unit. I would assume you'll need a second person to help guide it out.

[xt2005bonbon]

This is what I posted on your original thread. Maybe that other thread should be deleted.

-->

wow. Lots of stuff going on here. 

Two things to check:

1. is the timing belt installed properly? Count the number of teeth between pulleys and check what the FSM says. Also, make sure the spacing for the timing belt guides follows FSM spec. Too tight and the belt can jump a couple of teeth = engine running like crap or not running at all.

2. I would also check valve clearance. It is possible that some valves (especially exhaust valves) are too tight.

Also, pretty obvious but still: are white coil pack plugs installed to the front (cyl1/2) and black to the back (i.e Cyl 3/4)?

Do you recall if the misfiring problem happened in all situation, or just at warm idle, cold idle, any rpm range?

[jnstrong]

Prior to this current issue I checked the timing belt and it was good, timing was correct. Comment from another forum suggested checking again, experience with belt failure shortly after rebuild due to poor quality parts. 

I checked valve clearance some time back. And nothing was out of spec. 

I don't recall the connector color, but have replaced 2 for being broken. I can say they are in the correct location. 

The issue occurred upon startup and definitely persisted through all operating ranges (didn't get into boost for safety). 

I do need to verify the coil pack plugs don't have a loose connection.

Thanks!

[jnstrong]

I deleted the other thread to keep things tidy.

[xt2005bonbon]

OK. What are your long term fuel trims now?

Also, which plugs are you using? I have been using the NGK Laser Iridium Spark Plugs 7913 SILFR6A. Have always been reliable.

Edited November 8, 2023 by xt2005bonbon

[jnstrong]

13 minutes ago, xt2005bonbon said:

OK. What are your long term fuel trims now?

Also, which plugs are you using? I have been using the NGK Laser Iridium Spark Plugs 7913 SILFR6A. Have always been reliable.

Ill have to check data on my laptop at home to tell you what the trims were. There was some time between when I left off tuning and the major issue started where the trims drifted similar to before cleaning the injectors. I ordered new injectors then. I figured my cleaning wasn't sufficient and a poor spray pattern had come back. 

Damn that reminds me, when I pulled the old injectors out, #1 & 2 were missing the replacement plastic caps. Figured the engine ate them, I couldn't see them in the intake and didn't think a little plastic would cause any issues. Thats what prompted the original compression check.

I didn't write down which plugs, but they are whatever was in the car when purchased. I do need to get a new set of plugs to test, good catch. I forget sometimes the plugs can be a culprit (Problems coming from diesel work)

[xt2005bonbon]

The plugs I listed are the preferred OEM. I just bought two sets for my cars. I know you have a lot on your plate, but since you've been tuning and all, you should look into the BtSsm app to monitor things. Much easier than carrying a laptop around.

[tysparks81]

As xt2005bonbon mentioned I would start with the easiest things to check spark plugs and look at the coil packs. if you see any rust or separation in the metal plates replace the coil.  Also if the connectors for the coil packs are switched the car will run but it will run rough and throw a miss fire code so I might make sense just to double check that the correct plugs are in the correct place.

[Max Capacity]

White plug in front, black to rear.

 

[subisubisu]

6 hours ago, Max Capacity said:

White plug in front, black to rear.

AC/DC fans might remember it as "Black in Back".

[jnstrong]

1 hour ago, subisubisu said:

AC/DC fans might remember it as "Black in Back".

Ha Exactly what I thought of when I checked.

 

[jnstrong]

I picked up a leakdown tester and fuel service kit to get some better diagnostic info. Will report back with #s. New spark plugs are on their way.

Ive got suggestions from several threads on thigs to check, here are some results:
    - Checked each coil pack plug while engine was running, they are in the correct position - no change while wiggling connector or wire loom
    - All plugs and coil packs are in good visual condition. Contacts are clean.
    - LTFT (AF Learning) when running well post tuning, before misfires started: 
        ○ -3.4% under 6 g/s, 4.8% 6-15g/s, 0.7% over 15g/s (See Screen shot)
        ○ STFT and LTFT are plotted against MAF in chart "STFT & LTFT 1" mixed driving conditions

-LTFT from a quick log with Cyl 1 missfiring consistently, plotted in "AF Learning OL". This was only running in open loop.

 

[xt2005bonbon]

ok. so coil pack plugs are good. Fuel trims are fine. Only one cylinder is misfiring now or all of them still? What does the misfire count per cylinder look like at idle for each cylinder?

Also, how do your VVT angles look like? Are these at zero while idling and in sync under load? Is one lagging behind or completely off?

If you have already checked your valve clearance and are sure these are in specs, then your leakdown test should only confirm/deny something wrong with the rings. At least that's my logic. I am assuming you are going to do a warm leakdown test?

 

Oh, I also remember some guys who had their timing belt installed correctly, but it was running like crap. Culprit was at least one of the intake cam pulleys was not installed right, where it was misaligned  with the camshaft dowel pin. As a result, timing was off and a bunch of other problems.

Edited November 11, 2023 by xt2005bonbon

[jnstrong]

Update: Issue Resolved! It was an injector and tune issue. Full details below for those interested. Thank you all for the help.

I got new plugs in  (The original ones were correct as well but fouled) got those installed and fired it up. Engine was still misfiring on multiple cylinders. #4 was not firing, #1&2 had intermittent misfires. AFR was very rich. 
Warmed it up enough to get warm compression and leakdown tests done. Compression results were low across the board, but leakdown was good. Worst cylinder (#3, the only one not misfiring) had 12% leakage, the remainder were less than 10%. 

When this issues started, it was just #1 not firing. I had pulled stock injectors, when I put the engine back together I used the new injectors I had on hand (750cc). I reflashed the ecm with corrected injector flow and latency(or so I thought). I took a step back and went back to basics, checking every little thing. CEL was on, no surprise, but upon checking codes I noticed one I had turned off with the tune (P0038). Checked the image I created and it was all stock, no modified tables. Something happened when saving and no changes were saved! I should have checked, but haven’t encountered the error before so got complacent.

So no wonder it was super rich inj flow and latency was all wrong. I correct the image and re-flashed. Started the car, #4 was not firing and the AFR was way lean! Bingo #4 injector was the culprit.

I found when I installed the injectors I gorilla handed the connector back on and bent the pin. It was at least an easy fix, injector was saved and it runs normally again! Horay!! Now back to Tuning.
 

[xt2005bonbon]

Nice work on your troubleshooting. So, what are you exactly doing in terms of tuning?

[jnstrong]

Finishing installing the walbro 255 fuel pump (I now realize there are better options, but oh well) and finishing up the vac line for the fuel pressure reg. Moving to referencing off the intake plennum vs cyl4. 

For tuning, I need to retune the MAF and injector scaling (VF52 and 750cc inj). Been using the MAF tool in Romraider. Then on to the rest (Boost,POL fuel, timing, etc) Im in Denver and drive up to the mountains so I want to build a thorough and reliable tune. I may fine tune it on a dyno after first pass. But right now, just road tuning slowly. Its a side project and Im enjoying learning some new skills.

[xt2005bonbon]

Please post what you are doing regarding injector scaling. This is something I have yet to 'play with', but found a very good video from HP tuner last night, using our platform (well, an STI) and ECUflash.

 

-->

 

Edited November 22, 2023 by xt2005bonbon