M52TU Repeatable Cylinder 3 Misfire at Idle

[deadparrot]

My 1999 520i Touring (2.0 M52TU) is exhibiting an odd tendency to misfire on cylinder 3 after idling for a couple of minutes. Evidently the ECU is detecting an intermittent misfire >1/200 revolutions and switching off the injector as disconnecting the cylinder 3 coil has no effect once the misfire has set in. However, when the ignition is switched off and the car started again, it runs smoothly for a few minutes before the misfire sets in again.

So far I've tried swapping both coils and plugs with cylinder 1, which had no effect. Nor did disconnecting the MAF. The car does suffer from excessive oil consumption, which is suspected to be valve guide seals, and has had the CCV simplified to a basic PCV system.

My next thought is to try swapping the injectors around, but will I need replacement O-Rings before doing this? I expect the injectors have been in there for the best part of 20 years!

Are there any other avenues of common failure worth pursuing? Ignition coil drivers? Sensors?

[JimLev]

If it runs OK when you start and then restart it the coil and injector drivers should be OK.
Swap the injectors. O-ring should be OK, mine are 18 years old and still pliable.

[deadparrot]

If it runs OK when you start and then restart it the coil and injector drivers should be OK.
Swap the injectors. O-ring should be OK, mine are 18 years old and still pliable.

Thanks for the confidence in the O-Rings, I'll give injector swapping a go in the week.

I've also come across reports that it is typical for the M52TU/M54 heads to crack near cylinder 3, causing coolant to enter the combustion chamber and cause misfires. This engine did experience a massive overheating event which was sufficient to boil the coolant within the radiator and cause it to expand like an accordion! Before I read this I had largely attributed the smoking exhaust to the oil consumption issue, but now I'm inclined to believe that it's possibly a coolant leak relating to the misfire...

Aside from the smoking exhaust I've not seen any of other typical symptoms such as emulsified oil or oil in the coolant - is there anything else I can look out for to prove/disprove the cracked head theory?

[JimLev]

You could do a leak down test on the cylinders as well as a pressure test on the cooling system.
Another possibility could be the intake manifold gaskets.