[mictang]

So I am trying to chase down a rough idle, already replaced/checked all the obvious things:

• injectors tested in ISTA
• new NGK OE plugs
• latest Delphi coils, and swapped them around cylinders
• PCV hoses OK
• DISA valves checked and #2 were replaced
• compression is good and consistent across cylinders
• 4 new bosch OE O2 sensors
• ESS plug is dry
• new VC and gasket

The roughness reported in INPA is mostly in bank 2 or sometimes in cylinder 6.

At one point found the HVAs a little noisy and there was rather minimal play when compressed on a couple of them, replaced all of them, and there was a very small improvement (to be fair, could have been within "margin of error" as well).

Visually checked the VANOS adjusters while I was there and bolts were not backed out, the shop did use timing tool to reinstall everything. (Pretty sure the bearing ledge is okay too, didn't look super closely but no visual wear)

The thing is, there does not seem to be a way to isolate either VANOS or Valvetronic, if I unplug any of: eccentric shaft sensor, Valvetronic motor, VANOS solenoid, both VANOS and Valvetronic goes into fail-safe mode (VANOS in resting position, max valve lift), and the idle smoothes out (not 100% perfect but good enough).

Then, I read somewhere (https://bimmerprofs.com/rough-run-vanos/) that VANOS actual values were not supposed to fluctuate much.

But my intake VANOS actual values fluctuates ~8 degrees at idle and so does its PWM control values, and this is after replacing the solenoids with Pierberg ones (OE supplier), before which there had been an intermittent 2A82 code, freeze frame shows set point 79 degrees and actual 115 at ~1100 rpm, does not throw a check engine light.

Exhaust side seems pretty okay, actual value only fluctuates ~1 degree near setpoint, did swap solenoids too and fluctuation does not follow solenoid.

Did check non-return valve (very clean, only passes air in one direction) and oil filter cage (it's there).

I haven't checked oil pressure since I don't have a gauge, but I also think it would affect both intake and exhaust?

So, right now my primary suspicion is the intake side VANOS adjuster itself, since I have already ruled out solenoid, non-return valve, and engine oil filter, as well as the fact the fluctuation only happens on one side.

Or are there any other VANOS-related parts I could check? Would you guys recommend checking the oil pressure anyway?

I guess timing issue could be a possibility but that would require opening up the valve cover right? Then I might as well replace the intake VANOS adjuster.

Also, as mentioned earlier, is there a way to rule out Valvetrnoic issues more definitively?

[rjahl]

A few degree of the camshaft will not cause a rough idle. How far was the VAnos from the target value?

Try pulling the MAF sensor. See if it idles better without it.

.

[mictang]

Originally Posted by rjahl

A few degree of the camshaft will not cause a rough idle. How far was the VAnos from the target value?

Try pulling the MAF sensor. See if it idles better without it.

.

At idle, intake target is 84 and actual value is fluctuating between 79-88.

the behavior looks pretty similar to the second video in this link (~00:30 mark), where in their case was a solenoid problem in the exhaust side: https://bimmerprofs.com/rough-run-vanos/



pulling MAF sensor makes little (worse) to no change, I did check MAF values and they look pretty normal (just below 20 KG/H, AC on, ~690-730 rpm), so was lamada (idles at .98-1)

[WheelNut2]

I'm going through a similar struggle with my 330i. I've checked most of the same things you have (except for the coils) and I'm suspecting my problem could be the same as yours. Please keep us updated! I'm going to check my Vanos actual vs set angles and see if there is a similar discrepancy.

I like your diagnosis methods- seems pretty rigorous 👍

[mictang]

Originally Posted by WheelNut2

I'm going through a similar struggle with my 330i. I've checked most of the same things you have (except for the coils) and I'm suspecting my problem could be the same as yours. Please keep us updated! I'm going to check my Vanos actual vs set angles and see if there is a similar discrepancy.

I like your diagnosis methods- seems pretty rigorous 👍

I just went over to read your post, my rough running values are pretty similar to yours as well, but in bank 2 (funny right?).

The codes you are getting after resetting adaptions are normal (because the throttle has lost its adaption values due to the reset), I usually just have to clear the codes once or twice.



Back to the idle issue itself, from what I can gather, Valvetronic and VANOS play a pretty significant role in idle quality control, BMW once specifically claimed they improve idle smoothness on N series engines.

I have no noticeable drivability problem which further strengthens my suspicion of them causing the idle issue.

When Valvetronic has a code or if manually set to max lift (unplug the connecter and turning the hex screw on the vanos motor counter-clockwise until slight resistance is felt, can also be verified in INPA/ISTA), car falls back to old-fashioned throttle-controlled idle, and it improves idle in my case.

But as I mentioned earlier, there is no obvious way to isolate this from VANOS, any fault in Valvetronic causes VANOS to fall back to resting positions (solenoids un-energized), while any fault in VANOS also sends Valvetronic to max lift.

Please do let me know what your VANOS values look like though, would love to have something to compare to. It might also be worth checking the eccentric shaft angle at idle, normal value is high 30s degrees IIRC.

[Jros]

My car does this too but only at cold start up idle, for the first ~30 seconds.

Disconnecting the valvetronic sensor makes idle perfectly smooth.

Noticed that fuel trims (short term) change a lot when in throttle body mode at idle, car seems to have less correction to do.

My car always had negative trims at idle (around -6/-8 total); in fall back throttle body air control mode they fall back in the 0+-2 % range.

Never bothered me much, i suspect it is due to tolerances no more perfect in the valvetronic system, which has to mantain very tight lift (under 1mm) at idle.

Maybe related, i think the rough start got worse when i installed (with specific tune) MILVS

[Jros]

on a side note: this is the third N52 engine i have in this car: 2 N52B25 and this one, a N52B30.

They all more or less had the same behaviour so i think it is an N52 / Valvetronic II "quirk"(had MILVS in the last 2 engines)

[rjahl]

Originally Posted by Jros

on a side note: this is the third N52 engine i have in this car: 2 N52B25 and this one, a N52B30.

They all more or less had the same behaviour so i think it is an N52 / Valvetronic II "quirk"(had MILVS in the last 2 engines)

Are you running a custom tune for the MILVS? Have you tried flashing the stock tunes to see if the MILVS tune is part of the problem.

[mictang]

Originally Posted by Jros

on a side note: this is the third N52 engine i have in this car: 2 N52B25 and this one, a N52B30.

They all more or less had the same behavior so i think it is an N52 / Valvetronic II "quirk"(had MILVS in the last 2 engines)

Interestingly enough, I too have an N52B30 swapped euro 325i, how's your 22RPD tune? I thought about getting it but the rough idle really bugs me as it is permanent unlike yours only at cold start.

Right now I am just running one of the stock 30i calibrations. (S8608461.0DA, downloaded from bimmerlabs using a MSV80 330i vin)

My original B25 idled perfectly with zero vibration, until it lost compression in one of the cylinders that is.

If you don't mind, could you take a look at your VANOS angles when the car is warm? When the engine first went in I had always suspected Valvetronic, but after replacing the HVAs and with VANOS seemingly having an "issue", I am just a little paranoid about what really is causing the idle problem. If the few degrees of VANOS fluctuations are normal then I might dig into Valvetronic again, had this car for quite a few years and would love to keep running it as long as I can.

[Jros]

Originally Posted by rjahl

Are you running a custom tune for the MILVS? Have you tried flashing the stock tunes to see if the MILVS tune is part of the problem.

running custom tune by 22RPD, rough cold idle was there even with stock tune, USDM or EUDM tunes were the same in this respect.

I think reinstalling stock vavetronic supports would be the only way to pinpoint MILVS as the cause, not going to do that

[Jros]

Originally Posted by mictang

Interestingly enough, I too have an N52B30 swapped euro 325i, how's your 22RPD tune? I thought about getting it but the rough idle really bugs me as it is permanent unlike yours only at cold start.

Right now I am just running one of the stock 30i calibrations. (S8608461.0DA, downloaded from bimmerlabs using a MSV80 330i vin)

My original B25 idled perfectly with zero vibration, until it lost compression in one of the cylinders that is.

If you don't mind, could you take a look at your VANOS angles when the car is warm? When the engine first went in I had always suspected Valvetronic, but after replacing the HVAs and with VANOS seemingly having an "issue", I am just a little paranoid about what really is causing the idle problem. If the few degrees of VANOS fluctuations are normal then I might dig into Valvetronic again, had this car for quite a few years and would love to keep running it as long as I can.

could you share via PM the VIN you used? i was searching for a way to do just that to compare Bimmerlabs / BMW vs 22RPD

My first b25 cracked the block and i never understood why, that sucked...my second one blown rings in cylinders 2 and 3 without any warning.

that b*stard used like 1 liter of oil every 50 Km...so i got fed up and put a B30 in the car.

As for 22RPD, my actual tune works well, but required many iteration to get good, my base software comes from a USDM 328i from what i know, that may have been a factor.

The people at 22RPD is very helpful and fast reacting to any request, I can reccomend them.

as for logs, at the moment i have not a specific log tailored on vanos requested vs actual; i have this cold start with cam positions (all 3) logged though, with many other parameters:

https://datazap.me/u/jros/v6-vanos?l...4&zoom=250-404

Please take note my target idle is 750 and not 650; in the first seconds of the log you can see the car almost stalls and immediately after gets smooth apparently without any explanation.

Take note that i have no cold start routine too, so idle doesn't stay high for a minute but immediately targets 750 rpm.

If I find the time i may do a log more tailored to the specific situation.

[Jros]

and here, at the end of the same log, idle with a warm engine:

https://datazap.me/u/jros/v6-vanos?l...zoom=5783-5827

[TGTIW]

Originally Posted by Jros

My car does this too but only at cold start up idle, for the first ~30 seconds.

Disconnecting the valvetronic sensor makes idle perfectly smooth.

Noticed that fuel trims (short term) change a lot when in throttle body mode at idle, car seems to have less correction to do.

My car always had negative trims at idle (around -6/-8 total); in fall back throttle body air control mode they fall back in the 0+-2 % range.

Never bothered me much, i suspect it is due to tolerances no more perfect in the valvetronic system, which has to mantain very tight lift (under 1mm) at idle.

Maybe related, i think the rough start got worse when i installed (with specific tune) MILVS

I'm having the sames issue. Rough and surging idle for the first 30sec of a cold start, dead smooth after that. Like yours, mine also worsened after installing milvs and a tune (Stage FP in my case)

[mictang]

Originally Posted by Jros

could you share via PM the VIN you used? i was searching for a way to do just that to compare Bimmerlabs / BMW vs 22RPD

My first b25 cracked the block and i never understood why, that sucked...my second one blown rings in cylinders 2 and 3 without any warning.

that b*stard used like 1 liter of oil every 50 Km...so i got fed up and put a B30 in the car.

As for 22RPD, my actual tune works well, but required many iteration to get good, my base software comes from a USDM 328i from what i know, that may have been a factor.

The people at 22RPD is very helpful and fast reacting to any request, I can reccomend them.

as for logs, at the moment i have not a specific log tailored on vanos requested vs actual; i have this cold start with cam positions (all 3) logged though, with many other parameters:

https://datazap.me/u/jros/v6-vanos?l...4&zoom=250-404

Please take note my target idle is 750 and not 650; in the first seconds of the log you can see the car almost stalls and immediately after gets smooth apparently without any explanation.

Take note that i have no cold start routine too, so idle doesn't stay high for a minute but immediately targets 750 rpm.

If I find the time i may do a log more tailored to the specific situation.

If your firmware is 0049QK0MI20S, you should be able to just flash this file, otherwise let me know and I'll try to dig up that VIN on my phone, it has been a few months.

E92-330-Stock-S8608461.0da.zip

[mictang]

Originally Posted by Jros

and here, at the end of the same log, idle with a warm engine:

https://datazap.me/u/jros/v6-vanos?l...zoom=5783-5827

Looks like your VVT angles are lower than stock due to the MILVs and exhaust camshaft angle was changed by the tune?

I downloaded your csv file and did some calculations on the last 80 lines:

average exhaust camshaft 4.965189948 standard deviation 1.438918583
average intake camshaft -36.05696203 standard deviation 0.255388615

(noticed it wasn't fully warm, ~85C coolant temp)

mine at ~100C coolant temp, 300 lines:

intake vanos avg 84.1±2.49, camshaft -35.9±2.5
exhaust vanos avg 106.77±0.33 camshaft 7.9±0.23

my stock VVT is 27.6 avg, setpoint and actual.


The 2A82 code (intake vanos) came back, and considering my larger deviations in the intake, I am leaning toward vanos issue. (when the code is active, car idled pretty smooth with vanos and VVT in emergency mode)

Weirdly enough, in one of my previous logs, the intake camshaft values were almost identical to yours at ~36±~0.25. Guess I'll have to buy an oil pressure gauge just to rule things out.

[mictang]

And here are some visualizations of my funky intake fluctuations (the peaks are~ 10 seconds apart) compared to Jros' datalog:

[Jros]

when i drive somewhat aggressively my car always stays around 85 celsius coolant temp. I suppose that is due to the mapped thermostat doing its job but i would not be surprised if it (the thermostat) is dying instead but no codes at the moment

[Jros]

regarding the sawtooth shaped fluctuations, they could be something not holding hydraulic pressure, but this is a really out there guess.

oil filter cage, intake solenoid, actuator bolts falling out or worse bearing ledge?

EDIT: i see you checked or replaced almost all in the vanos system apart from actuators themselves; never heard of a bad actuator apart from bolts but it would be my main suspect at this point, or as you said oil pressure

[mictang]

Originally Posted by Jros

regarding the sawtooth shaped fluctuations, they could be something not holding hydraulic pressure, but this is a really out there guess.

oil filter cage, intake solenoid, actuator bolts falling out or worse bearing ledge?

EDIT: i see you checked or replaced almost all in the vanos system apart from actuators themselves; never heard of a bad actuator apart from bolts but it would be my main suspect at this point, or as you said oil pressure

I can confidently rule out vanos actuator bolts, solenoids, and oil filter cage. Looked at ISTA service info again and seems they only call for visual inspection for grooving/excessive wear on the bearing ledges so as far as I can tell they are okay too. (The engine was pulled from a low mileage late model car anyway.)

Also checked VANOS setpoint values against actual today, looks like the fluctuations were just happening by themselves, setpoint value is pretty much a straight line at idle.

I have already ordered an oil pressure gauge and will update on the results.


EDIT:
Found an old log from May that was taken after replacing the HVAs, VANOS values were smooth at idle (exhaust side wasn't perfect, ~3 degrees of fluctuation but there weren't any sawtooth shapes in the line chart, intake was rather steady at ~84 degrees).

Earlier than that, the car was actually throwing both 2A82 and 2A87(exhaust) codes, 2A82 was happening at a lower frequency, 2A87 freeze frame was like -115 actual vs. -102 setpoint at idle.

Replaced the solenoids (non-OE, but reputable brand, and replaced them again more recently with OE ones) before doing the HVAs, haven't had 2A87 since, VANOS behavior at the time was the same as the log in May.

Since the intake VANOS actuator was removed and reinstalled while doing the HVAs, this somewhat rules out installation error as it worked correctly afterwards. The most obvious variable from then to now is actually oil (fresh change back then), so it feels more like oil pressure issue...

[Jros]

Actually this engine is oil picky, but decreased or fluctuating oil pressure with a less then new / perfect oil would surprise me. But i was proven wrong many times in my life

[mictang]

Well, oil pressure seems fine, warm idle ~ 2 bar (Bentley manual calls for 1.5 but I am running Castrol 0w-40 which is relatively thick), max ~ 7.2 bar (also over the 6 bar max Bentley calls for). There wasn't much fluctuation in oil pressure either.

I don't think a bit higher oil pressure can lead to the behavior I was seeing in VANOS, it was literally like a cycle of losing a bit of pressure/angle and then compensates and overshoots the setpoint.

Did a bit of live data monitoring on the road today, intake VANOS seems to have brief periods of steady values when coming to a stop, and idle smoothes out for a few seconds (this behavior had been observed before).

Any ideas?

[Jros]

i could think about a cam bolt loosing torque and/or reluctor wheel losing position on intake side?

I don't know, if it is not electronic or hydraulic related i can only think about something mechanical being off, and thinking about the HVA job maybe it could be possible something was not torqued right or has lost the correct torque somehow, putting play in the mechanism.

the problem is that for checking this the valve cover must come off, so it would be better to check timing and intake cam bearing ledges (upper and lower) and to be sure change the bolts securing the vanos to the cams (one use torque + degree) at the same time...big job and not sure it would help

i don't think the software side can be at fault being and oem map.

swapping intake and exhaust cam position sensors changes nothing? if it does then the sensor now on intake side could be at fault