M42 Refuses to Run With AFM Plugged In

[recklss94]

Hello all!

Longtime viewer, first-time poster here on R3VLimited! As the title suggests, I’ve been having issues with the M42 in my 1994 BMW 318i. I’ll try to keep this as short as possible, but I tend to be pretty thorough—so I apologize in advance.

It’s worth mentioning that the car has RHD ITBs and has been running on the stock ECU for the past three years with basically no issues.

The Issue:
The car will not idle under any circumstances with the AFM plugged in. If I unplug it, the car dies every time. However, with the AFM unplugged, the car will run—albeit very rich.

1. Original Issue (Last Fall) – The car wouldn’t idle when warm.
2. Replaced AFM (Winter) – Installed a used AFM, but results were inconclusive at the time.
3. Original Issue Worsened (This Spring) – Once it warmed up, I confirmed the AFM replacement didn’t fix the issue. The problem gradually worsened until the car would no longer idle at all—cold or warm.
4. Replaced O2 Sensor – No effect.
5. Swapped Relays – Tested to see if the main relay was bad. After swapping, the car eventually wouldn’t turn over and threw a 1225 (Knock Sensor #1 – ground fault or DME short circuit) code. I did not see any obvious ground faults in engine bay, and the symptoms seemed inline with a bad DME, which I had experienced with my original years ago.
6. Replaced DME – The car was already on its second DME, which had a sketchy history at best. I replaced it with the proper red-label DME ending in 282. This cleared the CEL, and the car now reads 1444 (no issues). However, I’m back to point 3—no idle when cold, but it will idle with the AFM unplugged. The only CEL/code I get is when I unplug the AFM, reading 1215 (AFM not sensing correctly)—which makes sense since it’s unplugged. It’s worth mentioning that throughout this entire process, I’ve been testing both the old and new AFM at every step and getting similar results.
7. Tested Harness Extension – When I originally installed the ITBs, I had to extend the AFM harness. Suspecting bad solder joints, I cut the extension out and tested it. I verified continuity on all five pins/wires and confirmed minimal resistance (< 0.2 ohms).

Where I’m At Now:
I’m at a bit of a loss. I know some will suggest switching to a standalone ECU and utilizing Alpha-N instead of relying on the AFM. And believe me, I’m with you—I plan to go standalone when I do a built-M42 with the ITBs in the next year or so. But since the car ran well for so long on the stock DME, I’d really like to get it working again and enjoy it while I build the new engine on the stand.

If anyone has any ideas or has dealt with no-idle issues on the M42, I’d love to pick your brain and get some direction on what to check next. Thanks so much! :)

[Spyke]

I've never had a used AFM work without having it refurbished, personally. Bav Restoration is the man you need.

Have you checked your cold start valve (if equipped) as well? Sometime it won't fire, or won't stop firing as well which causes a lean or rich issue.

[recklss94]

Spyke, I apologize for the late response. Thank you for your input!

I've never had a used AFM work without having it refurbished, personally. Bav Restoration is the man you need.

I got in touch with Bav Restoration and am sending out one of my AFMs to get refurbished. Thank you for the tip!

Have you checked your cold start valve (if equipped) as well? Sometime it won't fire, or won't stop firing as well which causes a lean or rich issue.

I have not checked the ICV since installing the ITBs. I’ve always heard that it could be a culprit for idle issues, but in my 11 years of ownership, it has never been particularly dirty or clogged when I’ve removed it. Still, I suppose it warrants another look to be sure!

Is there any part of you that thinks something in the fuel system could be responsible? For example, if my injectors were clogged or failing, could that cause the idle issues? I’ve been through two fuel pumps on this car, and I think I can categorically rule that out as a cause based on their mode of failure.

[roguetoaster]

Unplug the TPS and try again with the AFM plugged in. Following that, unplug other components one at a time to rule them out.

Check timing agreement cam to crank. If it's out, and the #1 cam lobes are pointing up and towards each other (IIRC) you probably have a bad crank damper.

[e30davie]

How are you running ITBS with the AFM? made a common plenum that feeds the itbs through the afm?

yer id be inclined to unplug everything except the afm and see what happends. does the AFM test up ok with the resistance specs?

if the afm tests up ok, then for sure start looking externally. do you know the history of the fuel pump?

can you bypass all the idle valve stuff and just crack the itbs and get it to idle?

[recklss94]

Unplug the TPS and try again with the AFM plugged in. Following that, unplug other components one at a time to rule them out.

Check timing agreement cam to crank. If it's out, and the #1 cam lobes are pointing up and towards each other (IIRC) you probably have a bad crank damper.

I previously tested other components, such as the TPS and the coolant temperature sensor, but nothing besides unplugging the AFM forces the car into a sustained open-loop idle. Hmm, I’ll keep the harmonic balancer in mind—thank you!

[recklss94]

How are you running ITBS with the AFM? made a common plenum that feeds the itbs through the afm?

Yessir! The RHD Engineering ITBs come with a full carbon plenum, so I’m just using a silicone coupler to connect it to the AFM.

yer id be inclined to unplug everything except the afm and see what happends. does the AFM test up ok with the resistance specs?

I never got a chance to test the AFM’s resistance specs, so I guess we’ll be able to rule it out once I get it back from Bav Restoration!

if the afm tests up ok, then for sure start looking externally. do you know the history of the fuel pump?

Yes, I did—a new fuel pump was installed within the last five to eight years, and it has less than 20,000 miles on it.

can you bypass all the idle valve stuff and just crack the itbs and get it to idle?

I can, but the car doesn’t run well in open-loop—it backfires through the ITBs like an old carbureted car.

[KMT]

Apologies for the late reply, signed up to the forum just now.

Had a similar issue on my M42. However, this was on an E30 so I'm not 100% certain that my case applies for the E36 wiring loom.

Anyway, in my case the symptoms you describe were caused by a embarrassing mistake a made during an engine swap, where I connected two wires beneath the plastic cover on the firewall which are to be disconnected for cars equipped with catalytic converter.

[recklss94]

She Lives!

Apologies for the late reply, signed up to the forum just now.

Had a similar issue on my M42. However, this was on an E30 so I'm not 100% certain that my case applies for the E36 wiring loom.

Anyway, in my case the symptoms you describe were caused by a embarrassing mistake a made during an engine swap, where I connected two wires beneath the plastic cover on the firewall which are to be disconnected for cars equipped with catalytic converter.

Thank you for your response, KMT! I tried looking into this, but didn't immediately see anything, Additionally, I am on my original engine.

As stated above though, the car is fixed after almost a year of unsuccessful troubleshooting!

TLDR: it ended up being a vacuum leak. I ended up buying a fog machine off Amazon and using it as a vacuum/smoke tester. When I put it up to the intake, I saw a ton of fog coming out of the back of the plenum. One of the bolts that holds the plenum to the sandwich plate had worked itself loose and fallen out during some spirited driving. However, this wasn't the only leak that had been causing problems. The plenum had significant leakage around the circumferential seam where it meets the sandwich plate. It was only after sealing this seam with helicopter tape that the car is now back to running as it should. I don't recall there being any sort of seal for joining these two pieces on this kit for the ITBs, so I think all the vibration and heat cycling might have just widened any natural gaps in the seam over time. I should probably email Racehead Engineering and let them know this could be a more widespread issue within their product line. I’m sure some sort of a rubber seal would fix it permanently.

For anyone who cares/for future reference, here is everything I went checked in order since my last update here:

• Sent 1 of 2 AFMs to Bav Restoration --> will follow up
• Removed ICV --> looks perfectly clean, just like when I put it in. Not sure if they can fail…
• Reinstalled ICV --> no effect
• Redid wiring harness extension for AFM --> same effect as before when plugged in and unplugged, unplugging ICV has no effect either though?
• Swapped DME back to original --> started showing same problem code as before
• Swapped DME back to new one --> code gone
• Finally got a working AFM back from Bav Restoration (previously had 2 AFMs from Bav Restoriation damaged in shipping by USPS) --> same results as with old AFM
• Need to do spark plugs, vacuum/smoke test, clean injectors, and then cap and rotor
• Did spark plugs --> car runs smoother but no fix on idle
• Read about vacuum leak through broken fuel injector o-ring. Tested with carb clean and found leak that raised idle. Did upper and lower o-rings on all 4 fuel injectors --> no change
• Got fog machine and liquid to do smoke test --> huge leak coming from back of plenum. Turns out a bolt backed out from the plenum and sandwich plate. Massive leak coming from hole as well as where sandwich plate meets the plenum. Plugged with new bolt and sealed seam with helicopter tape. Fixed.

Thank you all again for the advice. I really appreciate it!