Z3MCoupe.com Forums

Go Back   Z3MCoupe.com Forums > M Coupé Technical > Technical Discussion
HOME Gallery Register Members List Calendar Search Today's Posts Mark Forums Read

Thread Tools Display Modes
Old 25-04-2009, 08:36 AM
V12s V12s is offline
Join Date: Jun 2007
Posts: 89
Default Did you ever get this sorted?

I'm enduring the same problem! Did you find the solution?


Been having a play around this weekend to try to eliminate a few things.

1) First of all I removed the electrical connection from the AFM. You don't actually get an ECU warning light by doing this unless you stop the car and restart it again with it disconnected, but what this does is force the ECU to substitute 'average' values for the AFM as it has no real data to go on. I wouldn't recomend driving the car with the AFM disconnected but as I only need to rev the car in neutral then it was useful to see what difference it made. The answer - none at all! This would lead me to believe that the AFM is in fact working. I did subsequently remove the sensor itself to inspect it and it did have a thin layer of oily dirt on the 'hot film' element which I cleaned off with carb cleaner, but unsurprisingly it made no difference back in the car.

2) My attention then turned to a vacuum leak as sugested by Mike. First of all I removed the hose connecting the brake servo to the inlet manifold and sucked on the end of the hose to see if the brake servo was leaking air at all..... no problem there. There are a few other pipes and valves on the S54 feeding into the inlet manifold on the engine side of the throttle flaps so I disconnected the main feed here and bunged the pipe with a rubber bung. This meant that the only way air could get passed the throttle flaps was via the idle control valve which should eliminate the majority of potential vacuum leaks. Revving the car in neutral saw exactly the same behaviour - an undershoot of revs before returning to a steady idle. The problem was probably elsewhere.

I left the car at that point and went in for the day - feeling pretty dejected!

On Sunday I decided to put the laptop on the OBDII port again and have a look at the data. I had done this a few weeks before and everything looked normal. There was a fault code stored for the AFM (despite no indicator being lit on the dash) which I kind of ignored and I went on to look at the lambda probe voltages. The thing that occured to me at this point was that the car didn't really exhibit the fluctuating idle when it was cold - it was then that I realised this seemed to coincide with the changover between the lambda probes being run open loop (i.e. ignored when the engine was cold) and closed loop (after about 5 mins of the engine running at idle). It would seem only when the lambda signal was being utilised that I had the problem.

So this did point to a lambda sensor fault despite the car just going through the MOT with perfect emissions. I took the car out for a drive with the laptop on the passenger seat to get everything fully up to temperature and then pulled into a layby to look at the lambda voltages. What I did notice was that one of the lambda probes (the car has two sets, a couple of probes before the cat and a couple after it, along with an EGT sensor) was swinging in voltage between 0 and about 0.7V whilst the other was steady at about 0.7-0.8V. Does anybody know if this is normal?

I still had the self inflicted error code for the AFM stored so i decided to clear the ECU and set off home. This is when I got the surprise - as soon as I cleared the ECU fault code (which actually resets all the learned stored data) the idle problem completely disappeared!!! Reving upto 3K rpm and back to idle was rock solid, whereas before it would very nearly stall.

All the way back home the car felt smoother but I did notice a couple of times that the ECU put the lamda measurement into open loop due to a 'system fault'??? This did not store a fault code or seem to make any difference to the car at all but seems a little odd?

So, I'm left with a problem that seems to have completetly disappeared after nearly 18 months of annoying me (despite the battery being off a couple of times which I thought reset the ECU anyway) and I'm still not really sure of the culprit - or if indeed it is gone for good. It looks like the most likely culprit is one or more of the oxygen (lambda) sensors but as the car has four of them, replacing them isn't going to be cheap and perhaps is not even neccessary... unless the problem returns.

UPDATE- Drove to work this morning and the problem is back. Joy.[/QUOTE]
Reply With Quote
Old 25-04-2009, 02:25 PM
V12s V12s is offline
Join Date: Jun 2007
Posts: 89

The new MAF which was installed seems to be the issue. I just tried the old one and it runs fine!? I will ask BMW to exchange the part and try it again.
I would still be interested to receive anyones input.
Ive got 5 days to make sure its sorted as the car's booked on the rolling road for mapping!
Reply With Quote
Old 01-05-2009, 12:32 PM
cragg711220's Avatar
cragg711220 cragg711220 is offline
Senior Member
Join Date: Oct 2007
Location: Milton Keynes
Posts: 310

yeah I'm also keen to hear if you got this problem sorted.
Imola S50 1999
Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

All times are GMT. The time now is 08:06 AM.

Powered by vBulletin® Version 3.8.8
Copyright ©2000 - 2018, vBulletin Solutions, Inc.