New build - Rough off idle

[NC325iC]

Long story short, I bought a $400 m42+g240 about 2 years ago. Rebuilt it and have been slowly installing it into my '74 2002. It's been really slow because we had a kid in the middle of the project...

Finally got it to the point that it should be driveable. It starts up and it idles pretty well, but anytime I crack open the throttle it stumbles and then usually recovers and comes up a few rpm. Crack it farther and repeat. If you try to rev it quickly it sounds like a miss or a backfire even. I am going to try to get a video on Tuesday but would like to get some suggestions.

So far I have:
Checked TPS - GOOD
Smoke test intake - some leakage around filter adapter, but otherwise good
Crank sensor is new, cam sensor is old

[NC325iC]

Ok got two videos today. It also seems to be hunting a bit at cold startup idle. Hunts for a couple minutes then stabilizes. The first video (20sec) is the hunting idle.

The second video shows the off idle issue. Near the end of the video you can hear what I think was a backfire or a miss of some sort.

Thanks for any help

https://youtube.com/shorts/Zgq39NRkOnI?feature=share

https://youtube.com/shorts/hgryDFhgDZ8?feature=share

[bmwman91]

What is the condition of the crank damper wheel? The rubber ring that connects the outer toothed portion (which the crank sensor uses to measure crank position) to the inner hub can fail and cause really erratic crank sensor readings. If you get down in there, you should be able to push and pull on the toothed part by hand and see basically no movement. If it is coming loose, then you need to replace the damper ($$$ unfortunately). There is no way to fix it, and if you try to run it hard with a failing damper wheel you can cause some really nasty misfires which can damage the engine.

Other things that can cause this are a failing air flow meter and damaged wiring at the TPS connector under the boot.

[NC325iC]

Thanks for the tips I will review those items next.

Ive been using this thread as a guide to check sensors.

http://www.m42club.com/forum/index.php?topic=17926.0

My afm is reading strange. I don't remember the exact details so I might get some of how it's supposed to work incorrect. Measuring the resistance across 2 terminals and operating the door by hand you should see the resistance decrease as the door opens. On mine I saw it start to decrease at the first portion of the door but then it would increase again. I assume this is not right, and I will probably try to locate a good afm.

[NC325iC]

Forgot to ask for clarification on TPS operation. The thread I linked above seems to indicate the e36 and e30 had different TPS units with the e30 using a idle/Wot switch style like on an m20 and the e36 using a potentiometer. Can anyone confirm this because I have the potentiometer type on my motor.

[bmwman91]

E30 M42 uses a throttle potentiometer, not a switch like the M20.

You cannot test the AFM by measuring resistance. As you noticed, the resistance is not monotonically changing as the door opens, and this is because of how the precision resistor taps on the carbon wiper track were laser trimmed to achieve the desired logarithmic voltage output function. You can test it if it is plugged in and the key is turned to accessory-II position (ECU has power). Pull back the boot on the plug and probe between the ground (grey/blue wire) and air flow signal (grey/yellow wire) pins. The voltage should vary smoothly as you open the AFM door.

[jsnppp]

bmwman91 I am constantly floored by your depth of knowledge on these cars. Kudos. How the heck did you know it is a logarithmic function?

Awesome, just awesome.

[bmwman91]

Haha, well I developed an AFM to MAF sensor conversion many years ago. It involved a bunch of embedded system programming and signal processing to do the sensor transfer function conversion. I was able to get the datasheets for several of the AFMs for old BMWs directly from Bosch through their field apps engineering support. Transfer functions worked out to perfect logarithmic functions. As for the MAFs, the transfer functions were much less easy to model with simple functions, and I used some specialized curve fitting software to get a good model with a closed form function. It was all totally overkill since linear interpolation between the provided air flow / voltage points would have been perfectly adequate too. But, where is the fun in “good enough”?!

[NC325iC]

bmwman91 I am constantly floored by your depth of knowledge on these cars. Kudos. How the heck did you know it is a logarithmic function?

Awesome, just awesome.

Agreed, total legend. Haven't gotten to check those things out yet but I appreciate the help.

[NC325iC]

What is this jumper harness between the afm and the main harness? I probed for voltage on the main engine harness between the two wires you mentioned (also confirmed in the etm). With the jumper I got a constant 3.6ish volts. Without the jumper I got a good sweep up to 5v.

Tried to run without the jumper and no change in operation.

I took a second look at the damper and no visible damage. I also don't recall any odd behavior from when I had it off the engine.

Just found some previous repair work to the TPS wiring. Will investigate now and let you guys know the results

[NC325iC]

TPS wiring is good. Checked it out at the ecu connection. Cracked open the ecu to look for any obvious damage and didn't find anything that looks cooked.

Is there another way to check the damper? The crank sensor is new, but it looks like a generic sensor, the wiring doesn't come off in the right direction... Maybe it's not the right sensor. Could be misaligned?

[NC325iC]

Here's a pic of the crank sensor

[NC325iC]

Back to bmwman91's prowess, you should see his thread about crank and cam position sensors on m42club... My searches have turned if up about a dozen times.

I did a bunch of searching for a maf conversion a couple years ago. Too bad your stuff didn't stick in commercial availability.

[NC325iC]

Looking up some parts. Is there a difference in operation of crank sensor 12141247259 (e36) and 12141721504 (e30)

Hmm just found this post in one of your old threads... I bought the Bosch one cause I'm cheap at heart.

https://www.r3vlimited.com/board/forum/e30-technical-forums/engine-drivetrain/m42/386826-m42-won-t-idle-some-backfiring-edit-photo-dump-of-broken-stuff?p=7064425#post7064425

[bmwman91]

Ah, yep you found the thread I was going to dig up. That should give you a decent idea of what to look out for. Basically, if you get a small pry bar or large flat head screwdriver between the timing case cover and the damper, the outer ring should not move axially at all when you pry lightly on it.

As for the crank sensor, I have no experience with the Bosch one...or at least not with the much-less-expensive Bosch one. The "Genuine BMW" sensor that I bought a few years ago was also made by Bosch, but it had the OEM style plug and the cable exited in the "proper" direction. A few people have definitely said that they had terrible issues with the cheaper one, which is really weird to me since one would think that the sensor's guts would be identical. But, maybe they are not. Enough things on them are different that I could see the sensors coming from 2 different production lines (OEM vs OE/aftermarket).

The E36 one probably just has a different cable length than the E30 one. The casting where it mounts on the engine is the same in either case.

So, I would say to swap it out for an OEM sensor, but these stupid things are pushing $300 now in some places lol. BTW, if you are buying Genuine BMW parts, I highly recommend this place. They almost always have the best prices and the customer service is awesome. Most of the hundreds of parts that I bought for my full repaint project came from them, and I have been a happy customer for many, many years.

https://www.getbmwparts.com/

But, before spending money on anything do a pry-test on that crank damper. What sort of repair do you see on the TPS sensor wiring? If it is under heat shrink tubing or tape, consider giving it a light tug-test to make sure that the repair is not coming apart.

The little interposer for the AFM is just a simple first order active low pass filter. Supposedly some M42's were having issues with noisy sensor output at idle or something, so this was a fix that BMW came up with. TBH the M42 does not really idle all that well on the stock ECU in any case, so I have no idea if the little filter actually does anything. I would just remove it.



Do you know when the last time was that the fuel filter was changed? Maybe that is clogged up. You may also want to get a fuel pressure tester on there to see if the fuel pump is consistently delivering pressure. You should be seeing ~45PSI above manifold pressure (so, 30PSI at idle with the engine running, 45PSi at WOT, and 45PSI if you just run the pump with the engine off).