[WheelNut2]

Hey Everybody,


TL;DR: Rough idle. Extra bad on hot start. Tried a bunch of things already, but what to check next?

Backstory:
I acquired an '06 330i about 4 months ago and I have been slowly chasing a rough idle and rough running problem ever since I got the car. This is my first 'modern' BMW. I owned a really beautiful E36 coupe before this car and a few Minis, but I've never had any car with so many computers on board before. I've setup ISTA, INPA, and I have an Autel OBDII scanner, so I have a number of tools to work with, but these things are new to me, so parsing out the data is a bit confusing. I've read many threads and watched tons of videos about misfires, and rough idle, but I haven't found the solution yet.


Symptoms:
1. Variable rough idle. Worst when hot starting. Mostly smooth at red lights. Cold idle is rough too, but maybe 20% better.
2. No codes!
3. Experienced what felt like a fuel cut at medium load on one drive. Three times on that drive, but it has never returned.
4. Some very slight misfiring when accelerating. Like a tiny stutter.

Investigations:
1. Replaced one VANOS solenoid. Swapped the two back and forth after. No effect.
2. New air filter. No effect (as expected. Old one wasn't too bad).
3. Cleaned MAF. Unplugged MAF and ran car. No effect.
4. Removed and cleaned Vanos check valve filters. They were clean. No effect.
5. Replaced all six spark plugs with new NGK IX Iridium LFR6AIX-11. Old plugs were a bit on the dark side, but didn't have significant build up and gap wasn't far off. No effect.
6. Checked battery age. "12/17" sticker present. Probably 2017. Probably okay? No weird voltage errors or codes popping up on ISTA.
7. Cleared all previous codes in DME. One Cyl 2 misfire code (29CE) was present, but I didn't note the date of it. It has not returned.
8. Smoke tested the intake using a home made smoke machine built out of a pickle jar, a soldering iron, mineral oil cotton balls and a bicycle tire pump. No leaks detected.
9. Used ISTA to individually de-activate fuel injectors one at a time. All de-activations caused mega rough running except Cyl 1. Re-activated all injectors and manually unplugged injector 1 and the car ran lumpy- you could hear it missing every revolution. I swapped injectors 1 & 2, re-installed and then re-tested with ISTA. Cyl 1 injector de-activation did nothing again. All other cylinder de-activations caused it to be run lumpy. Did the 9V battery test on the two injectors while they were out. They clicked when energized as they should.
10. Disconnected each coil pack individually and each one made the car run lumpy when disconnected, so it seems like they are working.

At this point I'm not really too sure what to look at next. I haven't put any fuel system cleaner through it yet. That's some low hanging fruit to deal with soon.

From what I read on Bimmerprofs it seem like maybe the O2 sensor readings are potentially wonky? Pre-cat lamba sensors are 1.00V and post cat lambdas are 0.75V. Wide band lambdas are 256ohm VK1&2, and 0ohm NK1&2. Does that seem right?

Should I move into investigating the various cam sensors and such? Is it possible to diagnose them with ISTA or INPA?

Thanks in advance for any help!

Attached are screen caps from the car at idle. Engine is warmed up and I believe the AC was on.

[k90]

My two cents:

Swap cyl1 ignition coil!!

If problem persists in cyl1 then rule out

Bad ignition voltage supply AND
Ignition coil connector fault AND
Wire harness fault AND
DME connector fault AND
DME transistor fault

Bad injector voltage supply AND
Injector connector fault AND
Wire harness fault AND
DME connector fault AND
DME transistor fault

AND

Check compressions

And then you could proceed to oscope position sensor signals for waveform anormalities.

All of them can also vary with temperature, but the position signal waveform anormalities are unlikely to vary with temperature.

Oxygen sensor causing just one cylinder problems sounds unlikelly.

[WheelNut2]

Interesting suggestions. I will try swapping coils.

k90 can you recommend a budget oscilloscope? How would one connect an oscilloscope to something like a camshaft position sensor?

[k90]

Originally Posted by WheelNut2

Interesting suggestions. I will try swapping coils.

k90 can you recommend a budget oscilloscope? How would one connect an oscilloscope to something like a camshaft position sensor?

ie. Picoscope 2channel will do.
A chinese pocket oscope if you have the time to go through tons of reviews and find a model that doesn't have flawed or missing functions or too small sample memory etc. And if you have time to wait it to travel across oceans.

Once did solder a thin wire to DME pcb bottom side connector solder joint for oscope probe hook adapter to attach to. Ground wire to pcb corner exposed ground IIRC. We actually found a problem then. Engine repair shop that rebuilt the engine had messed up the sheet metal crank position wheel. That engine barelly ran.
N52 doesn't have that problem with crank - the teeth are cut in the cast by the looks of it.

Now if I had to for sensor inputs to DME I would implement a branch wire between original sensor wire and DME. Because I have the pins and socket terminals to make such wire.

Did the coil swap help?

[WheelNut2]

I did the coil swap and it made no difference.

[Timo92]

I would check the mosfet if this one have them and rhe jnjectors.

[sabino56]

Just a thought as a long shot. Easy to check though maybe low likelihood. Not sure from discussion if it’s obvious misfire.

I also have a 2006 330i. I had a rough idle and knocking type sound which was driving me crazy. It turns out it was due a bad over running pulley on alternator. It was frozen solid and caused vibration in the belt/alt/crank. You could see it in the tensioner jumping. Also easy to check pulley on alt by hand with belt removed.
Thread on that discussion is here https://www.e90post.com/forums/showthread.php?t=1962637
Good luck!

[WheelNut2]

Interesting idea. My belt tensioner does move around a bit. Definitely more than it should be. Probably needs to be replaced. I imagine I'll be able to detension the belt easily and see how it smooth it spins. After looking at your thread I don't think my issue is the same. Your idle RPM speed seemed to be stable in the videos. My idle speed jumps around all over the place.

I might get another Vanos solenoid, or try swapping them again. I'm thinking to spray down the intake system with MAF cleaner too see if there is any RPM variation. Maybe my home brew smoke tester didn't pressurize enough to get any smoke leaking from the intake system.

I am going in for shoulder surgery next week, so I'll probably have to shelve this project for 3-5 weeks since I won't be able to drive or work on the car with only one arm! Hopefully I can do some more investigations this week before I become a temporary invalid.

[schwing]

Could be a fuel issue. I would run fuel injector cleaner as first step. Also, any gasoline smell behind driver's seat? Could be the fuel filter/fuel pressure regulator. Common issue with old, cracking plastic top which affects fuel pressure.

Replacement info
https://www.e90post.com/forums/showthread.php?t=1004917

[WheelNut2]

I did a bit more poking around this evening. I gave the MAF and its' connect a mega blasting of MAF cleaner and then did some logs of my fuel trims on a short drive around the neighborhood. I can't get my Autel unit to connect to my computer so I don't have any graphs to show, but STFTs were all in the -10 to +10% range as were the LTFTs. LTFTs were almost always negative on bank 1. LTFT1 seemed to bounce around between -10 and -2 while LTFT2 was pretty stable and slowly wobbled between -0.8 and -1.6. Cruising along in 3rd at about 50kmh (30mph) LTFT1 slowly settled at around -4. There was a spike on LTFT1 while off throttle of -15. Overall the engine ran pretty well as rough as before the MAF cleaning. Any effect from that, if any, was minor.

I also did a "leak test" that is purported to work by some folks: I blasted MAF cleaner all over the intake system and there were no RPM fluctuations.

There is no fuel smell in the cabin at all. I put a bottle of STP fuel injector cleaning into my tank last Tuesday and that tank is mostly burnt through now. I also have a bottle of Techron here for the next tank of fuel, so we'll see if either of those help at all.

[WheelNut2]

Small update. The car is running SLIGHTLY smoother now after the MAF cleaning. It isn't 100%, heck I wouldn't even call it 90%, but it is a little bit smoother now. Not sure how I could diagnose whether or not the MAF is actually working 100% correct or not though. The voltage signal from the MAF is analog, correct?

[sabino56]

I’ve no experience with it but I think it was M539 on YouTube I was watching and he said MAF can’t be cleaned to fix. There are places that can test, etc. you’d probably get all sorts of opinions on it with a Google? Good luck!

[WheelNut2]

I installed the second vanos solenoid a couple days ago. Rough idle is still present. I also spoke to a friend who has been a BMW tech for 15+ years. He said it sounds like a vacuum leak or vanos solenoids. I'll fix up my smoke tester a bit to hold pressure better and re-do my smoke test in a month or so when my left arm is back in action. Will keep you all updated.

[WheelNut2]

Update:
I was out of commission for nearly two months, but I'm back driving my 330i again. It is still running rough. Unfortunately parking it for six weeks wasn't a magical cure.

Did some investigating today regarding the manifold pressure. Had a look at the Bentley manual which has some specs for what the MAP sensor should be reading. From what I'm seeing here I think my manifold pressure is too high by about 20mbar. I was driving with the sunroof open today and noted that there was a perhaps bit of a whistle from the car as well. Maybe my CCV is done for?! What do you guys think? CCV problems are more pronounced at lower RPM, is that correct?

Attachments are:
1. Screen shot from Bentley manual with manifold pressures. I'm assuming the "operating range" is a pressure reading taken from the MAP sensor?
2. Scan tool showing MAP and ambient pressure at idle
3. Scan tool graph with MAP at idle and with engine at 2500rpm for a portion in the middle of the graph.

[Biginboca]

Check the PCV valve.

Your symptoms do align with the PCV being bad. Has the car been consuming oil? The N52 in my wife’s X5 was making the screeching sound when running and consuming like 1qt oil every 1000 miles. Those are some symptoms, others are rough idle and a lot of suction (excessive vacuum) on the oil cap when the motor is running.

[NZE90]

I have read the whole thread but didn't read anything about you testing the fuel pressure, (might have missed it though) It is supposed to be 75 psi. You will need a fuel pressure gauge to test it.

[WheelNut2]

Fuel pressure is shown as 603kpa (87psi) by INPA at idle ("Kraftstoffdruck"). There are no issues with smooth power delivery at high RPM and WOT, so I'd guess the fuel supply is adequate?

I ordered a new PCV and hoses. Should arrive next week. I'll report back after installation. I'll have a look at the DISA valves while I'm in there too.

[NZE90]

You can't measure fuel pressure with INPA. There is no low pressure fuel sensor on the N52, only N53 and N54 has a low pressure fuel sensor.

[chilledn52]

Shot in the dark here: pcv hose connecting from valve cover to oil separator possibly leaking and or internal crack in one of the pcv hoses?

[PhaseP]

Originally Posted by WheelNut2

Symptoms:
1. Variable rough idle. Worst when hot starting. Mostly smooth at red lights. Cold idle is rough too, but maybe 20% better.
2. No codes!
3. Experienced what felt like a fuel cut at medium load on one drive. Three times on that drive, but it has never returned.
4. Some very slight misfiring when accelerating. Like a tiny stutter.

When outside is low temps, it is showing air intake temperature (Ansauglufttemperatur) as 59 Centigrade, which seems high to me.

[lowrydr310]

Originally Posted by chilledn52

Shot in the dark here: pcv hose connecting from valve cover to oil separator possibly leaking and or internal crack in one of the pcv hoses?

I was about to say this. I had intermittent rough idle with my 330i that started randomly occurring, but would smooth out quickly. Eventually the SES light was triggered. I cleared the code and the same behavior continued. I rigged up a smoke tester and found the hose from the manifold to the oil separator was leaking.

[WheelNut2]

I removed the intake manifold on Monday. Plan was to replace the CCV unit and two of the three hoses, but it turns out that I ordered the wrong CCV and oil pan drain hose. New parts arrived today. All the hoses looked to be in perfect condition. There was some oil in the intake manifold- not a ton, but enough that 15-30mL (2 tablespoons) dripped out and there was a thin but complete oil layer covering the interior of the manifold. Disa valves seemed to be in great condition and the manifold and all parts looked to be in very good condition. New CCV will be going in this weekend and then I will report back on the results. Oh, also, I RTV'd the disa valve seals and I replaced the manifold to cylinder head gaskets.