Misfire/bucking under load solved: coil harness
 

This was a particularly difficult diagnosis for me, so I thought I'd share it. 2009 XC70 T6, 380,000 km or so. Symptoms: climbing hills or with light throttle at highway speeds the car would buck or shudder. It felt like a misfire, but there were no codes. On the highway it seemed to occur when fuel consumption exceeded approximately 11 L/100km. I still had decent acceleration if I put the pedal down. Idle was smooth. Bucking became severe on steep mountain grades. Starting took slightly more cranking than what I would consider normal. Tried logging misfires in VIDA, nothing. Changed plugs, swapped in a known good coil sequentially to see if the misfire would disappear. It didn't. Replaced all six coils anyway. Still no change. Cleaned the MAF (q-tip, alcohol - be careful!). Nope. Thinking the fuel pump or pump module might be going out, I logged fuel pressure (I had recently replaced the fuel pressure sensor in response to a sensor code). Fuel pressure was steady throughout rpm range. Logged boost, short term fuel trims, front O2 and rpms (see attached). Very weird. Whenever I felt the misfire, short term fuel trim would oscillate wildly, bouncing between -25% and +25% or so. There was also a slight oscillation in O2 and rpm (misfiring), but no misfires were logged. Oscillation in O2 was slight, but it tracked the oscillation in fuel trim. That made me think, maybe the computer was trying to adjust O2, but had to majorly overcompensate for some reason. Fuel restriction? I had the injectors cleaned and tested, no change. Spark plugs were all correctly gapped and in good condition - white insulators, no difference between cylinders. If anything the white colour suggested a lean condition. Compression was acceptable (135-145 psi on all cylinders), leak-down test was acceptable (it's a pain to do a leak-down test on these engines, as you have to remove the crank plug to bar them over). Smoke tested the intake, no leaks. Cleaned the throttle body. I had previously replaced the PCV diaphragm, and it was behaving as it should (slight vacuum at the dipstick at idle). Intake vacuum was rock solid, no needle flutter. The engine condition seemed very good, especially for a high mileage engine. I had also recently replaced the catalytic converter in response to a code, so no exhaust restriction.

Then I took a closer look at the logs. The misfire only occurred under heavy (for the stock tune) boost, around 16 psi. Spark is more difficult to generate under these conditions. Was this 'spark blowout'? Suggested an ignition issue, anyway. The coils were fine, I had swapped and replaced them. I checked the condenser (top of engine, black plastic thing connected between + and - of the ignition harness), it tested fine (DVM ohms test, starts low and runs to O/L). Engine run relay was good (swapped it for the starter relay), fuse box was good, no corrosion. Grounds were good, voltage drop between the coil ground and the battery negative was negligible. Back-probed the coils, system voltage (14v) at coil primary leads and the signal + leads. The coil harness was crusty and wires were cracked in spots but I have seen far worse. However, with nothing to lose I decided to repair the harness.

I haven't found Volvo coil connectors but aftermarket Toyota connectors are available with pre-crimped pigtails. The aftermarket connectors aren't great quality so I decided to purchase connectors with the longest wires available (NAPA Canada p/n UNI EC2438 - other suppliers carry the same part) but only use the wires/pins. I re-pinned the original Volvo coil connectors with the pins/wires from the aftermarket connectors, and spliced them into the harness using crimp connectors with heat shrink sleeves. The original harness has the 12v primary and the 12v signal + for the coils spliced in the harness above the valve cover. The splices are fused together and protected with some sort of sealant and heat shrink tubing. It seems that the wires in the 12v primary splice had begun to corrode, presumably restricting the current to the coils. I assume there was enough current to ignite a spark under normal conditions, but not enough to ignite a spark under load. In any case, the repaired harness has fixed the misfire. Hopefully this helps someone.
https://cimg1.ibsrv.net/gimg/volvofo...cc50d1bec5.png
Image shows logs of engine rpms, O2 and fuel trims going uphill at light throttle. Note the significant positive and negative oscillation in short term fuel trims. Long term fuel trims were normal.