For what it's worth, it was P0302 Cylinder 2 misfire detected (which corresopnded to the cylinder that had a plug boot dripping with oil)
So, here's where we've been since:
Diagnostic on catalytic converter showed excessive heating/blockage, (so perhaps the cat check engine light code wasn't completely worthless).
Dye test on head gasket showed a small internal breach. Shocking. Head gasket problems on a 2001 2.5L?
So we rebuild the heads both sides, replace the cat to the tune of $1800--New head gasket (aftermarket--wee!), new valve cover gaskets, seals etc, defect out the plug wires and plugs and all that jazz. Life should be swell right? Nahhhhhh.
Friggin stumbling/surge at 4000rpm persisted, but it was intermittent and generally only detectable after about 20min or so on the highway. Along with it, a mild bucking at 1250rpm on decel/coasting also became noticeable. So we limp around with that issue for a couple weeks since we want to give the rebuilt heads some breakin time and not rush to judgement. About 900 miles later, same crap is still happening and we finally get enough misfires to trigger another CEL code. This time: P0341 Camshaft position sensor circuit range/performance problem
So what the heck, we replace the cam sensor. $150 job. I get teh car back -- and no fix. Rough running in similar rpm ranges intermittently, and eventually the car did throw a check engine light (CEL). After sitting overnight in the garage though, before I coudl take it in, the CEL code cleared. Furthermore, when my mech went to read it, there was no code stored in memory this time.
At this point my mechanic and I are both frustrated with this headache PITA intermittent problem, and a nicely timed vacation allowed me to leave it with him to drive for 2 weeks with the Dx computer connected.
With the computer, he was able to graph cam and crank position sensor readings, and showed anomalies with these readings being out of sync when the misfires were occurring, and sometimes getting crankshaft sensor codes. During diagnosis and work, it was also noted that moving the actual wiring harness seemed to trip a code on at least one occasion. Continuity checks fine on all conductors though, so we have something intermittent--which as anyone who's tried to repair christmas lights knows, can be quite a pain. So, the last piece in that circuit is a common wiring harness that is shared by the crank and cam sensors. That part is due in today and will get replaced today. Meanwhile the crank sensor was replaced to rule it out (not on my dime though) as it did have some indications of damage on it in addition to the crankshaft codes that were being thrown (but not always remembered or stored after powering off the vehicle).
During this time, my mechanic enlisted the help of another independent master mechanic shop owner on this and got frustrated enough to call a local subaru dealer to bring it in on his dime for them to take a wack at it. After a long talk on the phone with their lead tech, he confirmed that these problems are frustrating and that he was doing all the same things they'd be doing, right down to having the same model of diagnostic computer they were using.
This car hates all of us right now.
So, I've had the car for the past few days after the crank sensor was replaced. Our stumble/surge at 4000 rpm on accel and 1250 on decel still persist. After enough of those misfires, I get a CEL and the codes I'm getting now, before the harness replacement: P0301....Cylinder 1 Misfire Detected P0302....Cylinder 2 Misfire Detected P0303....Cylinder 3 Misfire Detected P0304....Cylinder 4 Misfire Detected P0341....Camshaft Position Sensor 'A' Circuit Range/Performance (Bank1 or Single Sensor)
Fun fun. I'll report back after the wiring harness is replaced. If that doesn't fix it...... not sure if the ECU is just going nuts (that not remembering codes thing has me concerned) or what the hell is going on.
Best Regards,
-- Todd H.
2001 Legacy Outback Wagon, 2.5L H-4 Chicago, Illinois USA