Idle misfire diagnosis fun time: can exhaust leak/bad AF reading = single cyl miss?

[awfulwaffle]

Hey all,

 

As those of you wont to open this thread may have previously noticed, I've been hunting a #3 misfire at idle in my 2005 LGT for a while. Noticed that the misfire starts as soon as the TGV opens, but misfire counts drop to 0 if the motor is sufficiently hot (after hard driving, for example). I've replaced plugs, switched coils and injectors between cylinders and the miss stayed at #3. Replaced upper and lower TGV gaskets, no dice. Cleaned MAF and fixed a couple of boost leaks inherited from the previous owner, no change. Boost leak check with my janky DIY tester indicated all good, bike pump was able to ping up to 20 psi on the gauge without issue. When I attached a hose to the tester and tried to blow smoke into the engine, I quickly reached a point where I couldn't blow any harder (leaks sealed!). Performed (barely warm) compression test and checked valve clearance, and in both cases #3 was the best of all cylinders and all were in spec or very close to it.

 

Here's a picture of my cat chasing an F-15 to keep your attention and provide incentive to read further:

 

http://i92.photobucket.com/albums/l23/thesteakosaurus/FIGHTERCAT_zps5be2ff48.png

 

Realized today that I was interpreting AF learning backwards. I saw negative AF Learning at idle and interpreted that as the AF ratio being adjusted down, which would mean fuel was being added. This is what led me to think I might have a leak a vacuum leak at idle at the TGV.

 

Now that I know that negative AF Learning means reduction in injector pulse width, it seems that the car thinks it's running rich and pulling fuel. Clearly I don't have a boost leak at idle, so I can't figure out why this would be the case. I did take a good whiff around the engine bay today, and noticed that the passenger side of the engine does have a slight to medium exhaust smell. This smell is absent after hard driving. Now, this leads me to my question: could an exhaust leak somewhere at the passenger side of the engine cause a misfire in just one cylinder at idle? I know that the leak might cause a false AF sensor reading in that manifold and cause the car to adjust fueling for no reason, but just having a hard time reconciling why it would cause just one cylinder to miss. HALP!

 

I promise cupcakes and a tickle fight to whoever returns me my sanity.

[awfulwaffle]

Oh, forgot to mention my LV. Typically:

 

A: -12% to -8%

B: -2% to -0.5%

C: -4% to -10%

D: 0% as expected.

 

IAM is 1.

[awfulwaffle]

Anyone?

[Th3Franz]

Have you replaced the intake manifold gaskets ever? They can cause weird behavior when the car is cold. Known issue if you search around.

[awfulwaffle]

Yep, I replaced them last weekend. No change to speak of. In the process I discovered that the previous owner of my car somehow managed to leave the inlet and TB hose clams on finger tight, so my AF trims appear to be doing better now that the car's had some time to learn. Still, the #3 miss persists.

 

Also, turns out the exhaust smell was the fact that the clamp holding the turbo to the turbine housing was also barely tight. Currently going through everything I know to have been touched on the car -.-