Idle Surging (possible lean issue?)

[gespinal0301]

So I've been battling this issue for a while now. On a cold start the car idles fine, but once it warms up it starts going into a surge. Once I give it gas the car gets rid of the check engine light. On the highway the check engine light also comes on while cruising after a while and clears once I give it gas

Things I've checked/swapped out.

1) I upgraded to COP and got new plugs
2) Thought injectors could be clogged. Sent them out to be rebuilt same issue. (fuel pressure is good and installed a new fuel filter as well)
3) Checked for vacuum leaks (did MUIT about 3 or 4 years ago) couldn't find any
4) Replaced intake manifold gaskets as well as throttle body gasket

Heres a video of the car idling: https://www.dropbox.com/s/wp2zv4c55t..._3301.MOV?dl=0

Car running without AFM plugged in: https://www.dropbox.com/s/i5xl21niem..._3302.MOV?dl=0

Pics of the spark plugs are attached, not sure the cylinder order since I pulled these out a few days ago


Let me know if you guys have any ideas!

[roguetoaster]

Most likely it's a sensor with an intermittent fault, do basic checks on them, and read your CEL/look in to possible causes of that code. TPS certainly can do away with any throttle applied, probably won't come on during highway operation. While I doubt the plugs are contributing in any way to your issue(s), they're not the NGKs I'd suggest for a standard M42.

[gespinal0301]

Did some measurements, here's what I found:

TPS: measures 1.208k (0 throttle) -> 4.06k (full throttle)
CLT: 2.175k (Around 85* F in my garage)
IAT: 2.125k
AFM: 354.7 (closed) -> 134 (open)

With the AFM readings the measurements were jumping around a lot as I was moving the door flap. I saw resistance from 150 to 600 ohms. If I slowly opened the door the closed and opened readings would be different as well. I want to assume that this is the issue, but does anyone have measurements of their own I can compare to?

I wanted to test the Crank and Cam position sensors, but couldnt free them from their plugs. I tried pushing down the clips to free them but they wouldnt budge. Any tips on removing them?

[roguetoaster]

Don't have numbers to reference those provided against, but I assume you looked for specifications in the E36 Bentley or in one of the electronic troubleshooting manuals for an E30?

I am certain that the numbers have been listed here for the AFM, just a matter of finding them. also, I believe there are specs listed for the AFM when you check the above.

As to removing the connectors, and implement that will fit between the plug body and electrical connection box to allow you to pry them apart will work. They are often quite stuck in place. and the clips merely act as a safety retainer, so the clip must be pressed at the same time.

[gespinal0301]

According to Bentley the CLT temp spec are as follows:

Sensor Temperature(F)	Resistance
14	7000-11600
68	2100-2900
176	270-400

So mine is good to go.

IAT is stated to be between 2200 and 2700ohms so mine checks out

I couldn't find anything for the TPS or the AFM

I did measure my position sensors
Camshaft: 546
Crankshaft: 1291

Trying to find the spec for these in the mean time

[bmwman91]

The main thing that would affect idle between cold and warm is O2 sensor feedback. The ECU ignores it until the engine warms up to a certain point. Usually a crapped O2 leads to a rich, unstable idle, but maybe it could cause it to be unstable as well.

The coolant temp sensor is the only other one I can think of that would lead to this. Maybe it has a crack in the internal connections, and when it gets hot it expands and turns into an intermittent fault?

[gespinal0301]

What's considered warm? The issue pops up once my temp needle passes the very first mark, you can see it in the first video link

[bmwman91]

I forget the exact temperature when the O2 feedback is used, but it is warmer than that. I had not watched the videos until just now and it clears things up a little. When you said "surging" I had assumed far worse RPM deviations that that. What you have is just a mild "hunting" in the idle. That sort of thing is typically due to vacuum leaks and a gunked-up ICV. The cold after-start enrichments and ignition timing are all constrained a bit, and the target RPM is a tad higher, until the engine warms up a little (maybe about to the level you are observing) at which point it tries to more aggressively enforce idle speeds which can cause hunting behavior.

Other possible culprits are a worn carbon track in the AFM and maybe a worn TPS. The AFM does not have a resistance spec because its resistance is in no way linear as a function of door position. You check it by measuring output voltage. Try reading it at idle and see how much it fluctuates when cold-cold and when it warms up enough to hunt. You can also try reading the voltage while you reach in to the top of the air box and push the door open by hand. The voltage should increase smoothly from ~0.25V to ~4.8V. The TPS can be checked similarly. Move the throttl eby hand and see if the voltage output smoothly increases all the way up. You can just put the key in Accy-II mode as the Motronic will be powering and all of the sensors then.

I had a hunting idle like that a long time ago. I replaced the O2 sensor and cleaned the ICV to no avail. It turned out that there was a crack in my intake boot on the bottom which was really hard to see.

[gespinal0301]

I've searched for a leak using propane but I could never find anything. I'll try it out tomorrow and see if I can find it

[gespinal0301]

Does anyone have any good tips for finding a vacuum leak?

[roguetoaster]

Smoke works well, I think someone sells some in a spray can for those of us without a smoke machine.

Failing that, a visual inspection with disassembly should find it.

[ramerco]

OP, did you ever get this worked out? I have hunted hard for vac leaks without finding any and I have VERY similar behavior in my M20. I am trying to troubleshoot the O2 sensor most recently, and also wondering if one of my fuel injectors has a fault. But have tested (and tried multiple parts) the crank position sensor, TPS, AFM/VAM, and ICV. This is very irritating.

[gespinal0301]

Sorry life got busy for me and I've neglecting my car for a bit. I have not been able to find the issue yet. My next steps are to test the AFM and TPS, although I don't suspect it would be the TPS since I've replaced it once before. After that initial assessment I'll throw on a new O2 sensor and see if that helps any

[gespinal0301]

bmwman91 I've read somewhere that you can feed the AFM a square wave and read the output on o-scope for a more concrete test, is this true? Would reading the voltage range with a dmm be sufficient?

[bmwman91]

bmwman91 I've read somewhere that you can feed the AFM a square wave and read the output on o-scope for a more concrete test, is this true? Would reading the voltage range with a dmm be sufficient?

I can't say that I have heard of feeding a signal INTO the AFM, but there are some things you can do with a DMM. Due to how the carbon resistor track was designed, you can't measure resistance across the wiper as you move the flap since doing so will just give jumpy values. However, you should read a smoothly changing voltage. This would be done by manually opening the AFM door by hand while the car is in accessory-II mode, with the AFM plugged in and the boot pulled back from the plug. Stick the DMM probes across terminals 2 (gray / yellow wire) & 5 (gray / blue wire) to measure the voltage. It should be something like 0.25V when sitting closed, and increase smoothly as you push the flap door open, up to maybe 4.8V or so. The 2-3V range is probably where excessive wear would occur, so make sure it is smoothly varying there.

I've found it to be a hassle to open the flap by hand unless I remove the AFM from the upper filter box housing, so you might want to try to remove it from there. At the least, loosen the hose clamp and turn it upside down so it is easier to reach into the square opening if you leave the filter housing on.

The other thing you can do is to carefully pry the black plastic top off of the AFM and get a look at the moving bits in there. Mainly you would be looking for a deep groove worn into the black carbin track. IIRC you can radially shift the wiper arm a little so that it contacts a fresh part of the track, if needed. Re-seal the top lid with some RTV adhesive that is listed as "O2 Sensor Safe" or something like that.

See page 1360-4 in the link for the relevant diagram.

https://www.armchair.mb.ca/~dave/BMW/e30/e30_92.pdf