Head scratcher on M54B25 engine

[Krikkit]

I'm a bit puzzled on this one - have just bought a 325Ti Compact, no service history but ran OK so I'm giving it a good run through.

Over the weekend I sorted the cam cover seals as it was leaking like the Torrey Canyon, rebuilt the VANOS with new seals, and put 6x fresh plugs in.

The old plugs were quite old (being stamped with BMW they could be 10-15 years old!), but the same spec as the new ones.

After putting it all back together it ran OK on a cold start, but once warm it started misfiring. Diag laptop out, both lambda sensors were reading rich (ETA: pre-cat lambdas, one on each bank), and checking the MAF it was reporting an idle flow-rate of 9g/s - other owners have reported that usual idle flow is 3-4g/s, so it wouldn't surprise it was misfiring having had double the fuel stuck in.

As the car got hotter this got worse, I got stuck in traffic while giving it a quick test and eventually it just cut out. Unplug the MAF and it runs like a dream on the default map, no issues at all.

My question is, with the MAF not being touched, why on earth would old plugs and potentially a very slight timing change (from the VANOS) mask a rich-running issue?

Edited by Krikkit on Monday 7th October 11:16

[GreenV8S]

Maybe old wiring didn't like being disturbed and you now have an electrical fault on the MAF circuit?

[Krikkit]

Nope, checks out fine according to the multimeter... I've got another MAF on the way so we'll see what that says.

I can always put the old plugs back in I suppose

[twibs]

ecu coolant temperature sensor ok ?

[stevieturbo]

misfires on a narrowband will appear lean.

You need to find out why the MAF is reporting over double what it should....but if the NB sensors are reporting rich...clearly it is rich, and no wonder if the MAF is reading so high.

[Krikkit]

Thanks all, clearly the MAF isn't a red herring! If it was reporting lean I'd be looking for vacuum leaks etc.

Temperature sensors seem to be reporting sensibly.

I've run it for a few miles with the MAF disconnected and the codes cleared, the only ones present are MAF disconnected and a lambda heat on the post cat.

What I don't get is why a plug change would highlight it, especially on an NA motor with the same spec of plug.

Edited by Krikkit on Monday 7th October 21:10

[stevieturbo]

highlight what though ?

were you checking all these same things before ? what was the MAF reading before ?

Have you disturbed anything when you did the work ?

Was lambda reading rich before ?

[GreenV8S]

Is the fuel pressure regulator vac reference line connected OK?

[OnTheEdge]

Is the VANOS definitely working correctly after changing the seals? A sticking actuator or solenoid valve can cause rough idle etc.

[Krikkit]

Just wanted to bump this up with a resolution... Or so I hope!

Looked at the intake pipes and found that they were quite nasty, one had a tiny split in it. Replaced with new, but the errors didn't immediately disappear, but it did improve the idle quality.

Today I changed the fuel filter (which has integrated FPR) and connected the vacuum line back up (which was blocked off at both ends before) and it seems fine, and the error has disappeared. Only done a few miles but it's not reappeared, so hopefully now sorted.