Fuel injector replacement, startup issues

I spent the winter working on the car (2004 C4S, 101k miles): - New UAOS - I rebuilt the starter - 6 new fuel injectors (new Bosch) - Replaced most all of the vacuum lines - New alternator - Changed the oil (DT40) So, I had both intake manifolds, all of the vacuum system apart, and the fuel manifolds drained. On first start I had to crank it several times, but the fuel was depressurized, so I think that was normal. After a couple of cranks, the oil pressure was good (4-5bar) and the car finally started up. It ran pretty rough. I turned it off after a few seconds, checked for leaks, didn't see any. Restarted a couple of times. It starts strong, idles at 800rpm. After 20 seconds it starts to run really rough, and I get a cylinder 6 misfire in Durametric both times. (I'm monitoring those live.) Rough Running sensors are between -5 to +6, which is really high, before this they've only ever been around 0-1. I looked, and as far as I can tell, the electric plugs are on all the injectors. Any thoughts before I take it back apart to check injector 6 more closely?

Is it possible the injectors on that side are not fully seated?

Quotes source post: 17343440

I can't rule anything out, but installing, I remember them snapping down in, and the rail bolted down correctly, with the correct torque. So far no misfire for 4,5. I'm going to get my boroscope camea in there and I'll see if I can tell how their seated.

Car has been sitting a while, maybe pull that plug and see if its soaked with fuel, while you're in there you could swap the coil to another cylinder and see if the failure follows. Quick thing to rule out.

I'm getting ready to do exactly this over the summer. I''ll be following this thread. Best,

Quotes source post: 17343687

That is a good idea. I replaced all plugs and coils about 18 months ago, so they're fairly new, but that is a good trouble shooting step. I don't have much time to work on the car now (hence why it's April and I'm just now trying to get it started!), but I will follow everyone's advice as soon as I can and post with updates whenever I make progress. Thanks!

I'll try to keep a tally of what's been suggested, and what I've tried. Suggestions so far: 1) Check that the injector is seated properly. ( see images below ) : They look okay to me I think. 2) Check spark plug and swap coils to another cylinder. ( Pending... ) 2b) Scope cylinder ( My addition, Pending... ) 3) Pull injector 6 and test with 9V battery ( My addition, Pending.... ) Cylinder 4 Cylinder5 Cylinder 6 New and old injectors before installation, bank2

They look pretty well seated to me from the pix. You'd think if they weren't seated all the way you would smell or see gas.

Check for vacuum leaks at the oil fill tube and air intake. I've been burned at both items in the past.

Quotes source post: 17347518

Thanks. I haven't had time to do any work, but that does mean that I've had time to sit and think, and I was thinking the same thing. I did replace the oil fill tube while I was in there, as well as having so much else apart. I also realized that the Durametric was only reporting 1 single misfire in cylinder 6, not a continual state of misfires in cylinder 6. That is, the Actual Value for cylinder 6 misfires isn't a binary (0 or 1), but a total aggregate count. So for both startup tests, after about 30-45 seconds of running, it was really only counting 1 single misfire. It only ever showed 1. But the engine was running really rough overall. So, like you suggest, I'm now thinking that there is likely a vacuum/intake leak somewhere. But I hope to get some expert weigh-in on the next appropriate troubleshooting steps. Troubleshooting assuming it's an intake / vacuum leak: 1) Visually check intake and what I can see of the vacuum system (including the intake, manifolds, oil fill tube, AOS, AOS cross-over tube, etc.) 2) Try to start again, and record Fuel Trim values, record Pre-Cat O2 values. (?? Anything else? Can I hurt the engine by doing this for a little while? ??) 3) Smoke machine to test intake and vacuum system. Troubleshooting if it's only a misfire in cylinder 6: (I now think unlikely, if I'm reading the Durametric Actual Values correctly, so I'm waiting on these.) 1) Check that the injector is seated properly. ( See Post 7 ) : They look okay to me I think. 2) Check spark plug and swap coils to another cylinder. ( Pending...) 2b) Scope cylinder ( Pending... ) 3) Pull injector 6 and test with 9V battery ( Pending.... )

Update. Still looking for advice. Work/Family is making this the slowest troubleshooting process ever, but I'll get there! See data collected below where I recorded while it ran for about 50 seconds. The injector timing is all over the place right after it tries to go closed loop. Also, It's not one specific injector issue. Initially, the misfire was in Cyl 6. Then on startup this time, there was one in Cyl 5. In the data I recorded here, it registered a misfire in Cyl 2. So it's jumping round, just meaning that there is an overall issue with air/fuel. As you can see from the RPM, it runs really rough. (Oil pressure is good.) Troubleshooting assuming it's an intake / vacuum leak: 1) Visually check intake and what I can see of the vacuum system. --- DONE. I don't see anything obvious. 2) Try to start again, and record Fuel Trim values, record Pre-Cat O2 values. -- DONE. See data below. (I may have to get different O2 values) 3) Smoke machine to test intake and vacuum system. -- PENDING. Doing this in the next couple of days. Thoughts?

Im guessing the smoke machine will show u a leak. Probably a plenum boot.

Quotes source post: 17392886

Those plenum boots are a bitch... Smoke test can be done with a cigar and a transfer pump ($10 from advanced...). Hoping it points you in the right direction.

Thanks for the update. I'm grateful that the 911 is a second car and I can work on it over time. My plight will be very much like what you're doing while I do other things. It's a lot more fun when you don't have to drive the car on monday!

Did you have the new injectors flow tested before installing?

Check the fuel pressure at the rails.

I had a dead injector in my 996 Turbo. It ran VERY ROUGHLY and showed misfires on all cylinders. Interestingly it showed more misfires on the dead cylinder than any other. Watching the pre-cat O2 sensors I found bank 1 was running about stoich and bank 2 was really lean. The dead injector was on #5 and that's why bank 2 was so lean - no fuel going to one of the cylinders. I ran it for a minute and touched each exhaust pipe adjacent to the cylinders and found #5 cold, then inspected the plug which was Ok, swapped coil packs, etc. Finally I used a 12V battery to click each injector and found #5 did nothing. You can click the injectors from one of the big plugs on the right side connection points, just look up the wiring diagram. New injectors and it's back to normal.

SOLVED. Thank you everyone for your advice and input. I'm really glad I didn't have to pull injectors back out! It was an intake leak at the connection shown below. On the positive side, it was easy to get to. On the negative side, it was right in front of me, and I'm an idiot. There was a huge leak at the yellow joint. I had left it loose to position everything, and forgot to tighten it. So much ado about nothing. The orange joint was fit together correctly, but it is also leaking a small amount. I'll have to take this apart and try to find replacement O-rings. I'm pretty sure there are o-rings on both sides of the elbow joint. I tightened up the screws connecting the elbow joint to the manifold (yellow circle), and now everything runs great! I was able to take the car out for a drive, and I'm very happy. I was worried there for a while. I'll post some data later tonight just to follow up on my earlier data. I used this smoke machine. Amazon Amazon With the throttle body off, I put a rubber glove over the opening, and fed the tube through a cut finger. With some tape to seal the finger, it worked great. Should have taken a picture, but didn't. Huge leak (yellow). Minor leak (orange)

WIN !!! congrats on getting it fixed ! Happy for ya bud.

Found a small leak at your orange circle via smoke tester last month. Wrapped it in electric tape for now.

Quotes source post: 17395401

That's hilarious. Mine is also wrapped in electrical tape for now. I didn't post that; I was trying to look professional.

Not sure if it means anything to anyone at this point, but recorded some data after fixing the leak, and after a 40 minute drive. First start after fixing. It had idled for 30-60 seconds initially around 1100rpm. This is after the RPMs settled down. The battery had been disconnected, so the ECU is still adjusting. After a 40 minute drive. Idling. Fuel adaptations changed, but seem to be in okay ranges. It's interesting that the injector pulse is shorter, for the same RPMs. After a 40 minute drive. Idling. Just more baseline data. All rough running between -1 to 1.