I have a customer with a 2010 GMC Acadia.
Codes have been set for misfire on cylinders 1, 3, and 5 (essentially bank 1) and P0300 random/multiple misfire. Codes also set for P2097 (post cat fuel too rich). LT fuel trim is showing about -20%, more negative under load than at idle. Engine
runs rough with obvious misfire.
Historical data shows tens of thousands of misfires on cylinder 3, with about a thousand on 1 and 5. My thoughts are cylinder three is my true misfire and 1 and 5 are either registering wrong or being affected by the performance of cylinder three. I am able to command cylinder 3 off with my scan tool and the rate of misfires on cylinder three increase, so I do not believe cylinder 3 is a "dead cylinder", but it does misfire with great frequency. Cylinder 3 has good compression (~180 PSI) that matches it's neighboring cylinders.
02 sensors all show appropriate voltage and voltage gently moves between .1-.9 V as expected. All 02 sensors respond appropriately when I induce rich or lean condition with starter fluid or causing vacuum leak.
So, cylinder 3 misfire with tons of unburned fuel on that bank. Dealership changed spark plugs and recommended catalytic converter replacement (horrible idea, obviously cat inefficiency is result of misfire and raw fuel). I swapped ignition coils and plugs but the misfire did not change cylinders. Good spark present at cylinder 3 at spark plug. So...I replaced fuel injector on cylinder 3, with absolutely no change in any codes or live data. Also, the resistance on the old fuel injector was within spec and matched all other 5 injectors. I have perfect 60 PSI on fuel Schrader valve.
I read up some on direct fuel injectors and found a TSB for the intake valves being dirty on the intake side due to fuel not being present to "clean" the valves and causing misfires due to "tumbling of air". I cleaned the back of the intake valves thoroughly, but to no avail.
Reading my scan tool all of the bank 1 cylinders have a narrowed pulse injector voltage, as to be expected by the fuel trim numbers. Honestly, it seems to me like cylinder three fuel injector is just "dumping gas" like crazy, but the strange thing is I've replace it, and also the fact that I can command it off with a scan tool which makes me think wiring is at least (partially?) functioning.
My next thought it to backprobe cylinder 3 fuel injector with oscilloscope, capture what it looks like, and compare it to a backprobe of cylinder 1 or 5 and see if the suspect fuel injector (cylinder 3) is actually receiving the appropriate pulse width/voltage. If not, maybe wiring issue or PCM not actually sending appropriate signal? Am I missing something I should do first, because this is getting kind of drastic and my customer is unhappy with diagnostic costs (I don't feel like I've screwed the pooch or been sloppy in diagnosing, though).
I'm not very familiar with modern direct fuel injection, so I am hoping to find a GM guru to help me out with this one. I will pay for you to help enlighten me.