Off idle stumble - where to look next?

[dudecati]

Hi All - I'm chasing an off idle stumble my 91 318is has. It REALLY bothers me!

I believe I've eliminated all vacuum leaks with new hoses, I've replaced both intake gaskets, cleaned injectors, put in new air filter, fuel filter and plugs. Plug wires look good but I have ordered a COP kit to replace them anyway as the car has 188K on it and have no idea when/if they were replaced.

The car had a terrible stumble whenever I accelerated but new plugs (the wrong ones were in there) cured 98% of the issue.

Here are my current 'symptoms' - when car is cold it appears to idle and accelerate just fine (idle is smooth and right around 800 rpm). As it warms (temp gauge off the blue box to the middle range where it stays) is when the off idle stumble/bog occurs. It is just when pulling away in first and sometimes when going from off throttle to feathering it on in other gears. If I pull away hard it does not seem to be an issue. Something else I've noticed is that once warm, the idle climbs to 1000-1100 rpm but fairly steady (sometimes a quick stumble but nothing regular).

I'm thinking maybe an o2 sensor or the coolant temp sensor. Any thoughts if I'm going down the correct path? Is it possible to test those two items or is it just a replace and see what happens type job?

Any other places to look/advice would be appreciated too!

[DesktopDave]

I'd bet AFM resistance, dirty TB, flaky TPS.  O2 is usually a good bet too.  The DME doesn't pay much attention to those sensors until it's warmed up and running closed-loop.

You can test the AFM pretty easily with a multimeter just to eliminate it as a possibility.  Peel back the rubber boot & test pin 1 to ground - you should see 5VDC with the ignition on.  Between pins 1 & 2 on the AFM should show a gradual and moderate increase in resistance as you move the flap.  That's a tough test to do with the AFM still in the car.  Last, pins 4&5 should have a nice high resistance when air temp is cool (7-12K ohm if it's freezing) but that should drop as it gets warmer (2K ohm at 70degF).

Testing the O2 isn't easy - they only generate 1VDC at best, and only when they're really good and hot.