Yet another weird M42 stall problem

[mohaughn]

I've been trying to track down a weird stalling problem on my car for awhile now...  I've spent a lot of time going through recommendations on this site and a few others along with having a BMW master mechanic diagnose the problem.

The car- 1995 BMW 318ti with M42, 140k miles.  Not a daily driver.

The problem-  When the car is cold it will sometimes just cut out and stall.  It doesn't sputter, hesitate, etc..  It will just be idling and completely cut out, or if I'm moving and at a higher rev it will drop the revs down to 0 and sometimes the engine will come back to life.  Other times I have to move the key back to the OFF position, fire the ignition back up and it comes right back.  Once the car is warmed up I have never had the problem.  In fact I've run a 2 day DE at Sebring with this car since it started having this problem and it never once died on the track.

Whats be done-  The car is mechanically in great shape.  New fuel filter, fluids are frequently changed.  All vacuum hoses are fresh.  ICV and AFM have been removed and cleaned with FI safe intake cleaner.  O2 sensor has less than 35k miles on it since it was changed.  The camshaft position sensor is brand new.

For the past two days a BMW mechanic had the car and spent over 2 hours going over the car looking for any problems.  The car was checked for vacuum leaks and everything is good.  He checked fuel pressure and everything is good.  Car was put onto a diagnostic machine and everything reads fine.  The damn car never stalled once while my mechanic had it.  Last night after I got the car back it stalled on me.

So I called the mechanic back this morning to get more information on what he did and where to go from here and he mentioned that there was 1 code stored in the computer.  He did not have the exact fault code handy at the time I talked to him, be he said on the inspection sheet he noted it as AFM fault code that only applied to cars w/out catalytic converter.  He thought this was a bogus code as my car has a catalytic converter.

When I looked at realoem I see that there is an AFM listed as discontinued for M42's without a catalytic converter.  So I'm not really sure why my car is throwing a code for a car without a catalytic converter.  I do have a Turner motorsport chip in the car so it is possible that that is causing a weird code to be thrown.

So after that long post I have two questions.  Has anybody else ever had their US car with catalytic converter report an AFM fault that applies to cars without a catalytic converter?  I can't find any mention of anybody having this kind of problem.  Most M42 idle problems occur once the car is warmed up, not when it is cold.

And second, is a BMW part #13627558785 the same as a bosch 0 280 202 134?  Basically can any M42 AFM be used on the e36 m42?  I've found a used AFM for a decent price that I may pick up just to see if a new AFM clears the issue.

Got the code-

It is BMW Fault code 076 Idle co-potentiometer fault.  The GT1 gives what the correct value for this sensor should be, but says it is only installed on cars without a cat.  Is the GT1 and my mechanic on crack and this is really just the ICV going south?  I should also note that the five stomp method only gives 1444 over and over again.

[cecotto]

Since the problem only occurs when the engine is cold..

Could it be the engine temperature sensor thats gone bad.. If the engine management thinks it's at operating temperature, when it's stone cold it can have an influence on the running of the engine. then the engine will not have the warm up enrichment it needs.

Remember the old craburator days with manual choker, what this did was to limit air flow to the engine, this way effectivly enriching it. Modern engine mangement systems handle this enrichment electronically and relyes on correct inputs.....

Well it might be something else, but it' s worth a try...

[EN318isPDX]

Since the problem only occurs when the engine is cold..

Could it be the engine temperature sensor thats gone bad.. If the engine management thinks it's at operating temperature, when it's stone cold it can have an influence on the running of the engine. then the engine will not have the warm up enrichment it needs.

Remember the old craburator days with manual choker, what this did was to limit air flow to the engine, this way effectivly enriching it. Modern engine mangement systems handle this enrichment electronically and relyes on correct inputs.....

Well it might be something else, but it' s worth a try...

I had a similar issue with a 1977 Datsun 280z i rebuilt. After i got the motor in i couldn't get it to run right. Turned out after hours and hours of troubleshooting it was the temp sensor

[mohaughn]

So nobody has ever seen an "076 Idle CO Potentiometer" fault code on a US car that has a catalytic converter?  I've confirmed that I have the correct AFM, I've tried an AFM that I was able to get real cheap from a wrecked 318, and I still have the same issue.  So not AFM related as the bad code kind of implies.  

Temp sensors are cheap so that is always a thought, but I can't find any information on this fault code.

[Vladi]

Change the ICV...AFM doesn't influence your idle. :rolleyes:

[mohaughn]

If the car sputtered and hunted for idle I would have gone right for that.  The problem is that the car idles perfectly.  It will just ramdonly cut.  It will cut at 800rpms as easily as it will cut at 3,000rpm if the engine is cold.

I guess I did say in my first post that it will be idling along, so sorry for any confusion, but it has died while in gear, moving down the road at 45 MPH.  It will cut, and either stay dead until I recrank it, or it will come back from the transmission forcing the engine over.  If I push in the clutch once it dies, it will stay dead until recranked.

When I picked up this "new" AFM I was hoping to grab the ICV from the wrecked car as well but it had a different kind of ICV, L shaped instead of the T shaped one.

Since another diagnostic is going to cost me atleast another hour of labor I'm just going to replace the ICV and temp sensors for a a little bit more.  Was just hoping somebody on here had seen this code before and had a better idea of what could really be causing the problem and reporting this bogus code.

[Ramblin MAn]

When that happened to me, it was a bad hall effect generator (crank sensor) that went bad over several months. The car would just die all of a sudden. It was not temp related and was not an M42 so take it for what it's worth. You would assume you would get a different code.

I have seen where code numbers are "reused" so 076 may mean something totally different on a US car than CO potentiometer. You'd think a dealer mechanic would have that info, but i have rarely if ever been impressed with dealership mechanics.

[mohaughn]

Wanted to update this thread.  I found a TIS article about rouch idle issues on the M42.  It mentions that a faulty ICV can cause miscodes CELs in certain circumstances.  Based on the input from some people here I also decided to change out the engine temp sensor.

Changed both out last night and have run the car from a completely cold engine to normal operating temperate twice now without the car stalling.

I think it was the temp sensor that fixed it as the car runs cold now.  Before changing this sensor it would always run at the usual 12o'clock position.  Now if it is sitting idle it will go up to 12 o'clock, but once the car starts moving in 60 degree weather the temp drops down cold again.  I have a new thermostat to fix this problem, but it never did this on the old sensor.  Makes me think the car thought the engine was warmer than it was and was changing the enrichment at the wrong time causing it to stall/sputter.  Hopefully I don't have to update this again to say that I still have the problem.