Rough idle - driven mad

[Starachell]

Hello all. I'm currently suffering from the dreaded rough idle and I'm at a complete loss what to do next. I have a 2011 n52 2.5 manual which I purchased last July from an independent dealer. After being attracted to the car for its low mileage and pristine bodywork, it has now turned into the worst car I've owned. I've listed the main points here for brevity but it's worth noting that in between this issue, I've had to replace a bodged trailing arm, remove the adaptive suspension and a broken spring following an MOT fail, install a b12 kit, repair the solenoids on the trunk and change the battery. I say that to point out the rough idle issue has been there for a while but feels like it's getting progressively worse because I'm now driving the car more consistently whereas between Oct - Feb she was in and out of the mechanics for these other issues. 

Main codes pulled by our code reader and that of the specialist:
POOBC
2D06

Symptoms:
-The car will stall in neutral or in first gear when waiting to move off at a light/roundabout
-The car will also attempt to stall and then will rev itself back to life - this is less frequent
-The rev counter and temperature gauge bounce when the car is stationary. This rev hunting often pre-empts the above issues happening, though sometimes the car dies with no warning. The worst experience I've had was in nose to tail traffic for an hour and a half and the car either died, or dropped the revs and then revived itself, 16 times.
-Occasionally, upon restarting, the car seems to shake and whirr
-Having the air con on makes all of the above worse, particularly the shaking/whirring
-Heavy/rough handling below 3000 revs whereby the car feels sluggish and it takes a lot of work to get a response. For example, I might be in 4th gear doing between 30 -40mph and the car computer can be saying go up to 5th and I'll have to drop to 3rd to get some engagement and steady things. Interestingly, I feel this issue is worse when the engine is warm rather than at startup. I can start a journey well, get stuck at a long light, the car dies and then for the rest of the journey I will feel like I'm pushing the car to get home.
-I end up often dropping gear to increase the revs to eliminate that feeling that the car is about to die
-The car bunny hops at low speed/revs and at higher speed will almost skip a beat before resuming
-Increase in fuel consumption - no surprise given that I'm dropping gears
-Exhausts are coked
-edited to add: car now has EML but no new codes 
-Completely unrelated I'm sure but the alarm has now started to go off in the rain. I've seen the issue with the microsensors in the door card so we're going to assume that's a bonus gift. 

As I said above, this is all happening more frequently. However, I have days where I drive the car and she behaves herself and I'm left wondering if it's a figment of my imagination. Other days I look like I can't work a clutch. Typically, the good days are when she's on the ramp or being driven by a tech to find the issue! As a daily driver, it's getting to the point where I don't feel safe and I find myself taking different routes home to avoid long traffic lights or steep hills where I may need to stop for cyclists etc.

Here's where we're at with the diagnostics and fixes:
-I had a new MAF sensor fitted under warranty when the issue first started. I've since had that replaced just in case the part was faulty or the work wasn't done. This made no difference whatsoever. 
-We've cleaned the MAP sensor which seemed to improve things for 2 weeks but at this point, that could also be down to the moon being in the right position with a favorable wind 
-Smoke test didn't show any leaks
-We've just cleaned the Vanos solenoids with no real difference. The morning after the solenoids were cleaned, the rev needle seemed stable but by the time I drove home she was stalling again
-Battery changed
-The specialist's best guess is that the fuel pump needs changing but they have been very honest to say it's an educated guess and they're only about 50% sure this is the right direction. That assessment is down to the 2D06 code indicating a lean mix. With all the work done so far, I need more certainty than that. I would also expect some of the long crank issues. From what I've seen, fuel pump isn't high on the list of culprits for this issue. 

In my naive mind then I've got a list of potential suspects:
-Vanos solenoids need replacing and the cleaning wasn't enough, though they did whirr into action afterwards
-Fuel pump
-Sparks and coil packs
-DME 
-Oil pump as I've seen some people talk about the lack of a proper seal 
I know there are lots of posts about this here but I've struggled to find one that had the idle without clear misfire or long crank. I'd be immensely grateful if anyone could share a similar story - particularly if you have a positive outcome that doesn't start to stretch into the four figure territory.

[PAC_Z4_F31_110Puma]

If it were me and having had a rough idle and poor running (new solenoids sorted it), to stop chasing my tail with an undiagnosed falt, I'd make a fresh start and wipe the slate clean with a couple of cheaper end-of-the-spectrum options, first check closely for any splits in the intake side, make sure all the grounding points are tight for coils, then given the vanos is totally reliant upon oil, change the solenoids to new ones, I'd bell your local dealer - in my experience, they are actually price competitive and you know they're not a knock-off, especially given your experience of cleaning temporarily improving things. Do a fresh oil change with quality oil & OEM filter and new housing o rings, pull out & clean the vanos filters (can be a bit of a pig to get to), if you don't know the age of them pop new sparks in or at least pull them out to check the condition, especially given the poor running. At least from that point onwards diagnosis should become easier to pinpoint and rule out the basics, plus all good preventative stuff anyway. After that, O2 sensors would be my next point of query- a lot of trust is placed in the data being accurate. My line of work golden rule is to be methodical and skeptical, bad data in will always result in bad data out, and until you're sure of the basics, everything else is just guesswork.

[smorris_12]

Another vote for air leaks and/or lambda probes. There's an awful lot of vanos associated codes possible so, if none are being reported, I doubt it's that.

[Starachell]

Another vote for air leaks and/or lambda probes. There's an awful lot of vanos associated codes possible so, if none are being reported, I doubt it's that.

That's what I wondered about the codes. It's hard to know whether to start with the codes or the symptoms as they don't overlap. We were hoping the smoke test would've turned up something just to know what was going on.

[Starachell]

If it were me and having had a rough idle and poor running (new solenoids sorted it), to stop chasing my tail with an undiagnosed falt, I'd make a fresh start and wipe the slate clean with a couple of cheaper end-of-the-spectrum options,

Thank you, I think starting with the easier items is probably the best course of action before working our way up. I can't see any record of the solenoids being changed and the sparks were changed as per schedule but then the car only did 1,000 miles per year between 2019 and 2021 so I wonder if refreshing them would be wise in any case.

[AveZ4coop]

Hi OP, I had a similar issue 3 years ago. I had no codes, a rough idle, a distinct lack of power and looked like a 'clutch virgin' when pulling away from a standing start. It was very frustrating.

On the advice of someone on here (sorry I can't remember who as they deserve full credit!) I replaced coil packs and spark plugs and problem was solved. Has been running well ever since. Definately start with the cheap stuff first (check hoses, air leaks, spark plugs, coils, etc..).

I hope you get this resolved soon, these are great cars and I am sorry to hear you are not getting to experience that.

[Starachell]

Hi OP, I had a similar issue 3 years ago. I had no codes, a rough idle, a distinct lack of power and looked like a 'clutch virgin' when pulling away from a standing start. It was very frustrating.

Thank you, it is hugely frustrating! When you had this issue, did your car stall as well?

[matsmith749]

I would be seconding the coil packs / injectors as a likely culprit (not had the issue though myself).

A good starting point is to find the list (on here) of common issues - then see which of those could be at the root of your problem.

Coil packs / injectors are high failure rate items on these engines - I would be thinking about changing them out if they are original & take it from there.

[flybobbie]

Pull plugs to make sure no oil leaks into plug holes.
I had rough idle but cleared after 3-4 seconds.
Found oil in no.4 plug hole.
Fitted new cam cover and gasket, all sorted.

Failing that next would look at coils, just replace the lot and be done with.

[Starachell]

Thanks all for the ongoing insight. As an update, the engine light came on while driving home this evening for the first time ever. My specialist is kindly going to pull the codes in the morning and the car has now gone into limp mode. From bimmer link we're getting 2D06 - value below threshold with a note that the error would cause a warning indicator. Before, we didn't have the added indicator message. So I'm still scratching my head and wondering what might be underlying this. We'll be checking sparks and coils following your advice and the pins to the MAF to check they're seated correctly. We're also wondering if perhaps the continental MAF is the right part. Some sites suggest the part number on mine is for the 3L and could cause issues (clutching at straws now). I've seen lots of comments about the rough idle but not as many people saying the car hunts and then dies.

[deltasierra]

In general you may get one coil or injector malfunction causing misfiring this is much more general rough running affecting all cylinders, so air leaks, lambda sensors, crankshaft sensors etc the OBD scan will point you in the right direction, a very small air leak can cause havoc so check all the intake system thoroughly.

[Starachell]

In general you may get one coil or injector malfunction causing misfiring this is much more general rough running affecting all cylinders, so air leaks, lambda sensors, crankshaft sensors etc the OBD scan will point you in the right direction, a very small air leak can cause havoc so check all the intake system thoroughly.

This is it, the car isn't misfiring which is why we didn't do coil packs/sparks initially. No air leaks noticed following a smoke test so that seems to be well ruled out as well. The OBD scan points us back to the MAF but that's been changed twice so it's a gremlin somewhere else which is feeding into the MAF code.

[deltasierra]

I’m not a BMW expert but I do have an N52 engine in my Z4 and I like to know what makes it tick, nobody has mentioned “Valvetronic” (not Vanos) The airflow into the engine is not only controlled by MAF it is also controlled by valve opening, there is a stepper motor that controls the operation, I’ve never had to fully investigate but there is troubleshooting information online.

If you Google BMW Valvetronic it will explain the system

[matsmith749]

Googleing around for you - someone solved the 2d06 air MAF code by cleaning the oil cap!

Standard troubleshooting is to clean / replace the MAF, if fault remains you have air coming in from somewhere.

Might be worth checking / cleaning all the caps just in case it is something so simple as per the previous forum member - you could get lucky!

[colb]

Hope the Maf that was replaced was an OEM part, from experience cheap Mafs are a waste of time, they don't work or if they do dont last long, only fit oem Bosch or Seimens expensive but they work. Also any air leaks in the vaccum system across the engine from Maf to exhaust can cause havoc with fuel trims fooling the ecu into thinking its running lean and demanding more fuel at the injectors to richen the mixture up. This is due to unmetered air getting in and the exhaust sensors picking up the unmetered air that has not passed through the Maf and been measured. Splits or holes in any of the rubber vaccum hoses and pipes, intake boots and failed cam cover gaskets and central plug well gaskets can allow air leaks into the engine as well as oil leaks out. Have also seen a plastic cam cover degrading and develop cracks that resulted in vaccum leaks. Smoke tests usually find this sort of fault, arfter I found a leak on the inlet side and fixed that a subsequent repeat smoke test then found the bad cam cover leak. On that one a replacement cover was the only option with new gaskets.