N52 Headers with Fluctuating Idle

[onefastredsc]

I wanted to post this thread in hopes that it may help someone else down the line who comes across this same issue. Given how common the fluctuating idle issue is with N52s running AA or similar catless headers, it is all too easy to blame it on the headers or the tune. I am going to propose another failure point and how i diagnosed the issue.

When i first installed the AA headers, about 1 year ago now, the car initially seemed fine (i.e. idled fine, sounded a bit louder, had a little extra top end kick, etc.). But then after installing the StageFP tune, it immediately began having the infamous fluctuating idle problem. I then received some troubleshooting advice from Bob at StageFP to clear the adaptations and then take it for a drive. If the problem persists, the next step would have been to flash back to the stock tune in an effort to isolate the tune as the issue. However, after disconnecting the battery overnight to clear the DME's memory and then taking ~20 mile drive through back roads, the idle was smooth as silk. I have been running in this configuration for the last year trouble free.

Fast forward to a couple of weeks ago, and I had a local muffler shop (Meineke) install a N54 muffler on the car. I was searching for a little extra sound over the stock muffler and was impressed by the AA header plus 135i muffler combo i heard on another member's car. Since they welded the muffler onto the end of the axle back pipe, i suspect that they had to disconnect the battery. When i got the car back, the noise was perfect (and still is), but the fluctuating idle issue had returned. I tried resetting the adaptations again in hopes that a reset plus drive cycle would cure it as it did the last time, but this time the issue persisted.

My initial bias was to blame the headers and/or tune given the prevalence of posts complaining of fluctuating idle post headers install. I decided to drive the car with the protools app running, providing a live feed of short fuel trims and AFRs, both banks 1 and 2. The intent was to better understand why installing headers leads to the idle issue. I had initially speculated that perhaps a large AFR amplitude caused by locating the O2 sensors farther away from the exhaust ports was pissing off the DMEs feedback loop. However, what i found i did not expect. While bank 1 AFRs held relatively steady, which averaged right around stoich with an amplitude of approximately +-0.2, bank 2 was all over the place. When the idle issue was present, the reading would swing between 13 and 18, largely in phase with the fluctuations. Additionally, while out on the diagnostics drive, the cars idle would occasionally smoothen out and it appeared to correlate with times when the bank 2 O2 sensor would max out at ~44 and not provide any feedback. Note that while all of this was going on, the car did not trip a CEL.

After the above observations, i decided to do a little googling to understand O2 failures. I had always been under the assumption that an O2 sensor was a relatively binary device in terms of operating condition, i.e. it either worked or it didn't. However after a little searching two new revelations became clear; 1) O2 sensors can, indeed, progressively fail, and 2) they are quite delicate (i.e. sensitive to shock). I didn't necessarily throw them across my garage during the transfer to the new headers, but I definitely was not mindful in ensuring that they didn't impact the ground or other hard surfaces during the process either.

So i ordered two new primary (upstream) O2 sensors, installed them (carefully), cleared adaptations, and then went for a drive. The car is now once again silky smooth and both O2 sensors are reading normal (~14.7 -+0.2 while idling and driving around town). I hope this helps someone else who may be trying to diagnose a similar problem. Don't be too quick to blame the headers, tune, etc. until you have done some troubleshooting of the other components in the DMEs feedback loop.

[JemBash]

Originally Posted by onefastredsc

I wanted to post this thread in hopes that it may help someone else down the line who comes across this same issue. Given how common the fluctuating idle issue is with N52s running AA or similar catless headers, it is all too easy to blame it on the headers or the tune. I am going to propose another failure point and how i diagnosed the issue.

When i first installed the AA headers, about 1 year ago now, the car initially seemed fine (i.e. idled fine, sounded a bit louder, had a little extra top end kick, etc.). But then after installing the StageFP tune, it immediately began having the infamous fluctuating idle problem. I then received some troubleshooting advice from Bob at StageFP to clear the adaptations and then take it for a drive. If the problem persists, the next step would have been to flash back to the stock tune in an effort to isolate the tune as the issue. However, after disconnecting the battery overnight to clear the DME's memory and then taking ~20 mile drive through back roads, the idle was smooth as silk. I have been running in this configuration for the last year trouble free.

Fast forward to a couple of weeks ago, and I had a local muffler shop (Meineke) install a N54 muffler on the car. I was searching for a little extra sound over the stock muffler and was impressed by the AA header plus 135i muffler combo i heard on another member's car. Since they welded the muffler onto the end of the axle back pipe, i suspect that they had to disconnect the battery. When i got the car back, the noise was perfect (and still is), but the fluctuating idle issue had returned. I tried resetting the adaptations again in hopes that a reset plus drive cycle would cure it as it did the last time, but this time the issue persisted.

My initial bias was to blame the headers and/or tune given the prevalence of posts complaining of fluctuating idle post headers install. I decided to drive the car with the protools app running, providing a live feed of short fuel trims and AFRs, both banks 1 and 2. The intent was to better understand why installing headers leads to the idle issue. I had initially speculated that perhaps a large AFR amplitude caused by locating the O2 sensors farther away from the exhaust ports was pissing off the DMEs feedback loop. However, what i found i did not expect. While bank 1 AFRs held relatively steady, which averaged right around stoich with an amplitude of approximately +-0.2, bank 2 was all over the place. When the idle issue was present, the reading would swing between 13 and 18, largely in phase with the fluctuations. Additionally, while out on the diagnostics drive, the cars idle would occasionally smoothen out and it appeared to correlate with times when the bank 2 O2 sensor would max out at ~44 and not provide any feedback. Note that while all of this was going on, the car did not trip a CEL.

After the above observations, i decided to do a little googling to understand O2 failures. I had always been under the assumption that an O2 sensor was a relatively binary device in terms of operating condition, i.e. it either worked or it didn't. However after a little searching two new revelations became clear; 1) O2 sensors can, indeed, progressively fail, and 2) they are quite delicate (i.e. sensitive to shock). I didn't necessarily throw them across my garage during the transfer to the new headers, but I definitely was not mindful in ensuring that they didn't impact the ground or other hard surfaces during the process either.

So i ordered two new primary (upstream) O2 sensors, installed them (carefully), cleared adaptations, and then went for a drive. The car is now once again silky smooth and both O2 sensors are reading normal (~14.7 -+0.2 while idling and driving around town). I hope this helps someone else who may be trying to diagnose a similar problem. Don't be too quick to blame the headers, tune, etc. until you have done some troubleshooting of the other components in the DMEs feedback loop.

How you doing man I installed the AA headers 2 weeks ago and my tune still didn't come. But the car is running beautifully and not idling at all. I'm scared to install my tune tbh because it's working so well now

[onefastredsc]

Originally Posted by JemBash

How you doing man I installed the AA headers 2 weeks ago and my tune still didn't come. But the car is running beautifully and not idling at all. I'm scared to install my tune tbh because it's working so well now

If the car is running well post install on the factory tune, then i think it is safe to assume it'll be fine post tune. To be clear the intent of my post was to showcase how I blamed my issues initially on the tune, but in the end my fluctuating idle issue had nothing to do with the tune. The issue was with an O2 sensor that was on the way out and/or damaged during the installation of the headers.

If you have the 3SI manifold with the headers, get the tune. Definitely worth the smoother delivery of power (change in DISA switch points). If no 3SI manifold, then i would not bother spending the money on the tune (my opinion). My understanding of the control capabilities of the DME is such that the computer will constantly adjust it's control outputs based on the feedback it receives from it's inputs (sensors) in order to target ideal engine operation. As such, the DME will adjust to the revised operation of the engine due to the headers. In fact if you do a bit of searching, either here or on the E90 forum, running the catless headers without a tune is not uncommon. Only real downside is the tripping of the CEL due to a lack of catalyst efficiency detected. A proper tune will also hide the CEL if that is something that matters to you.