M42 Intermittent Problems

[downforce22]

So i have built an m42 powered 1984 318i lemons car. The car was a shell when I started with a roll cage so the wiring was created from scratch. The wiring was done to mimic oem wiring through a c101 with switched fuses. At last years race we had 1000 miles on the car and had nothing but problems at the race. Rust in the gas tank was the main problem. I changed the tank and did another test day so the car has over 2,000 miles on it now but at the recent test day kept shutting off and caused us to withdraw from the race next weekend. After 4 laps on track, the engine just shut off and I coasted to a shot to a stop. After a tow back to the pits, the car fired up right away. I went back out and it happened again. After 3 times the track wouldn't let me go back out on track so I worked on it and worked on it and had intermittent problems restarting the car and about 2 miles from the track it shut off and did not want to restart. After a 20 minute sit, it restarted and drove the 50 miles back from the track without problems. Then once home it wouldn't restart.


Here is the build thread: https://www.r3vlimited.com/board/for...e-car-with-m42

Symptoms/ things done:

- The fuel filter had some debris so I assumed it was a clogged fuel filter. Changed that out and on the next session during lap 2 the same thing happened.
- I had added a low pressure lift pump in the later tank to solve fuel starve. So I verified the lift pump was flowing to the 318is fuel pickup on the passenger side.
- Added 5 gallons of gas to full, just in case the lift pump wasn't working
- Bypassed the fuel pump relay and wired the pump directly to constant power and the engine still would not start the first try. (this tells me the problem is upstream of the chassis harness)
- had previously replaced the cps with a new oem CPS last fall
- had previously replaced coil on plugs with known good coils last fall
- swapped the ecu with 2 others and intermittently, it might start but might shut off while idling
- replaced both fuel pump relay and o2 relay and problem did not go away
- replaced the main relay with a good one and thought maybe that was bad and even bypassed/jumpered it with wiring and still the car shut off
- seems like the engine ground is solid, seems like connections at the battery are tight on power and ground.
- stomp test shows 1222 if the check engine light comes on and it is running a wideband oxygen sensor
- upgraded and replaced wiring form teh fuel pump relay c101 connection to the fuel pump and even wired the second pump on a second wire to avoid and pump problems with the wire gauge

Thoughts

So it appears possibly there are 2 or 3 more things it could be. At this point everything that seems logical has been changed out (ecu, afm, coils, oxygen sensor, main relay, fuel pump relay, fuel pump, fuel filter, etc.) It seems the car fires up and runs weel in the cold and does not do well when hot. This is distinctly different from the 'hot restart' afm wire problem which I found years ago on a different 318is and have not connected.

1 - There are a couple thread of the harmonic balancer separating the toothed wheel from the rubber. The one on the car seems solid. Any way to test this besides just separating the wheel? I do have a spare engine with one I could swap. But the car seems to run well then abruptly just shut off does not seem to be the likely cause.

2 - Wiring - The wiring was from scratch so I wonder if wiring upstream of the fuses is shorting out or something. I could check this wiring at the c101 or the wiring in the m42 engine harness for problems. The engine harness does not appear to have been modified. Anything downstream of the fuses has been working just fine.

3 - Grounds - Looking at the wiring diagrams, it appears the CPS, ECU and main relay are the most critical parts to starting the car. I have verified these three items but found this wiring diagram posted by bmwman91 which shows the m42 wiring in great detail. It appears that fuel is not the problem and it is coming from either the c101 or upstream of that. Looking at the diagram, I started wondering about G-103 ground junction connection near the battery tray. Last night I looked at it on the car and the bolt was kind of dirty and the cable itself was not very clean. Here is the interesting thing. With that ground disconnected, the car would still start and idle. I believe this may be the problem but need to do some driving around to test it and I think that means hot weather and hot conditions. Is there a better way to ground this wire than this screw hole at the battery tray? Directly the battery ground? I cleaned the terminal last night and when going to tighten, the bolt didn't seem to tighten all the way, like it was partially stripped or something. Its on the battery tray right behind that shock tower. You can almost see it here in these 2 photos. I think this is the problem but wanted to get other opinions. Since it is an m10 based car, the chassis is slightly different.

http://www.m42club.com/forum/index.php?topic=20060.0

Full wiring diagram here, I downloaded in case this link gets broken in the future.

http://www.e30tuner.com/assist/m42wireharness/EngineHarness_M42B18_E30_318iS_OEM.pdf

Does anybody have any thoughts on what I may be missing that could be the problem?

[bmwman91]

The battery tray should be perfectly fine for grounding, although I am not sure if it was set up exactly the same way in a 1984 chassis. Where does the battery itself connect to the chassis? Can you pull the battery and take a clear pic of the tray area & ground lugs?

If the 'G103' bolt is spinning in place then yeah that isn't ideal lol. As far as I can tell, 1984 models grounded stuff on the engine, so I am not sure what the setup looks like on the battery tray.

Your problem does sound like it is wiring-related. If you pulled the crank damper and the center hub is not separating from the outer ring, then it is probably fine.

[downforce22]

It has been a busy summer and the car is away from me. I was able to look at this past weekend. I removed the engine chassis ground and cleaned both sides of the connector. Next I removed the negative battery cable from chassis and engine oil pan, cleaned and retightened oil pan to chassis ground and verified no corrosion and retightened battery negative terminal ground to chassis. The started appears to be getting good power so I did not touch that cable. Then the battery tray ground was solid with a nut fastened on the bottom of the screw. The paint was ground off and this ground was also solid.

On the first 3 tries the car start up right away with no problems. It sat in the same spot in the garage, not having moved, and about 20 minutes later would not fire up. With nothing changed. It appears to be a wiring issue.

Then I bypassed the FPR and ran the fuel pump off a constant power fuse. The pump was buzzing and the engine would just crank. No start. That tells me it is not a fuel supply problem from the pump.

Next I disconnected the Groups coil plug and tested it and it was getting voltage. It seems the coils are getting power. I checked the resistance and voltage across the battery to the ground and they read the same with resistance very low.

Previously jumpered the main relay and the problem persisted.

At this stage there are 2 suspects. Either

the Harmonic balancer and crank sensor are not reading correctly and the ECU is not getting the CPS signal from the new CPS sensor
or
the engine harness has a problem that will sometimes overheat and interrupt the crank signal/coil signal/ecu ground/ ecu power/fuel injector signal.

My brother found this thread that says the following:

https://www.bimmerforums.com/forum/showthread.php?893452-M42-engine-won-t-start

For the no-start situation, i found that it was actually a bad connection... the wires underneath the lower intake manifold that connect to some sort of black box... the connector was the culprit.

That black box is where the crank position sensor and the cam position sensor plug into .Without Them the car will not run.

Sounds like your making progress and learning about your car at the same time.

good luck and keep up the good work.

So i have a spare engine harness and spare harmonic balancer and will be messing with those to test next. Is there any possibility that the CPS is located too far away from the harmonic balancer to get a good reading? It doesn't appear that is a possibility based on the bracket, but worth a thought of grinding down the mountiang bracket..

[Spyke]

Have you ohm'd out the crank sensor? Apparently a lot of new ones do not read the same values as the OEM ones, and they've caused a lot of issues on this engine in particular. I forget what the values are offhand, but you can find them in the forums fairly easily. It's also a shielded wire, so failed connectors can certainly cause this issue as well.

Coil connectors have a similar issue where a small break can cause them to not fire correctly. Another thing worth checking for drops in values, check it up to the ecu connector. There's a diagnostic cover on the ecu connector you can remove and backprobe each wire for the coils. then check it all the way to the coil connector, and look inside the boots there. I feel like while the m42 is not a powerhouse, it is a buzzy little engine that would greatly benefit from properly potted connectors on the coils.

[downforce22]

Thanks for the feedback Spyke, I will look up the CPS readings and verify, but this is an OEM bmw CPS replacement, not the unreliable replacement sensor.

With the m42 having coil packs and this car running coil on plug with working coils tested from a another car, I am confident the coils are not the main problem.

[Spyke]

Good luck man, this seems to be an almost immediate problem with nearly every COP converted M42 I've personally read about.

Not many are great at documenting either so we're often left with no real answers other than the things you've already been chasing. Damaged coil connector wiring, burnt up ECU's, shorting out ICV's and the crank sensors reading under OEM spec OHM's tend to always be the things I've seen or read about.

You seem like you're doing all the right things so I hope you find the culprit.

[Spyke]

***Found it, the correct value that the ECU likes for the crank sensor is roughly 550 Ohms. If you've got anything else, even the "good" BMW one, because at this point all of those are reproduced as well, toss it out.

[downforce22]

Closing the circle on this, The problem may have been solved. The CPS was not correctly clipped all the way into the connector for one. So When I clipped that in, it fired right up. After some more testing, the car shut off again and more probing, the main relay was not making a good connection with the the housing. If the housing was turned a certain direction it would shut off power to the ecu and car. This shouldn't be a common problem on most cars clipped into the relay holder on the firewall, but is a possibility.

[Spyke]

Damn, that's a weird one. Glad you tracked it down!