Help with Erratic Ignition & Check Engine Light

[510man]

I'm driving the '87 E30 Vert in the rain last night. It's running fine. Suddenly, it starts to buck. The gauge needles bounce. It runs like the ignition is cutting in and out. Idles poorly and won't accelerate without tremendous missing and cut out. Check Engine light comes on.

I limped the car about two blocks to a parking lot and left it over night given the rain. The next day, I started the car. It started easily. Same problem. Idles poorly, won't rev up and the Check Engine light is on. Tried multiple times to pull the codes but the car will not go into the fault code reader program. I've had this problem before.

After checking a bunch of connections, especially the CPS since I recently replaced it, I disconnected the battery to clear the codes. I put the battery cable back on and started the car. Runs perfectly!!!! No Check Engine light. No misfire. Remember, it would hardly run moments ago.

I'm stumped but need the car to be reliable. DMEs in these cars are a weak point. Mine is likely 7-8 years old with a Conforti chip. Should I replace it? Anybody used ecudoctors.com? Need to confirm this is the correct ECU number for my car.

http://www.ecudoctors.com/bmw-325-ecu-0261200173a.html.

Suggestions on what to try?

[603Racing]

I would diagnose my problem in more detail before throwing parts at it.

As for the ECU Doctors ECU's, that's expensive. If you're set on replacing it, do 10 minutes of research on this site on ECU's then buy a good used one from someone cheaper than ECU Doctors. Or borrow one from someone local to test. In fact, after a 30 second search I found this:
http://www.r3vlimited.com/board/show...&highlight=173

[510man]

Did it again today in the dry. Disconnected the battery to clear the codes and it runs perfectly again. Runs poorly until the codes are cleared.

Thx for the tip on the used ECU. I need to confirm I have a 173. ECU Doctors is a five year warranty so maybe worth the money. Records show this is the fifth replacement in this car or about every five years.

[jlevie]

Rarely does a DME need to be replaced. Five replacements suggests some problem with the car that is damaging the DME. Voltage spikes from a bad alternator, grounding issues, or a bad wiring harness are obvious suspects.

Or it could be that the prior replacements of the DME are the result of a misdiagnosis.

When power is removed from the DME, any stored fault codes are cleared along with learned adaptation. The clearing of the adaptation values may be why the engine runs better after power cycling the DME. But until the cause of the odd/invalid adaptation is found & fixed the cycle will repeat.

[510man]

DME checked out okay. Used a loaner DME and the same problem occurred. I believe I've eliminated the DME as the problem.

Disconnected the MAF sensor and the car ran well. Replaced the MAF. Progress but not fixed. In general, the car runs well. Has an occasional misfire. I believe it is ignition related given how erratic the remaining problem is and how it feels like a temporary ignition failure. The tach bounces when it misses too, further leading me to ignition. I would think a vacuum leak wouldn't cause the tach to bounce. No check engine light now.

What are the signs of a faulty CPS? I replaced mine but a new part doesn't mean it's good. I also used an aftermarket part. I've since seen some posts from others having problems with aftermarket CPS they replaced with a Bosch, as I recall, and problems where resolved. There's also the sensor on the ignition wire. Not sure what it does but it's clearly sensing timing for something.

Comments and thoughts on what to try next?

[jlevie]

The tachometer reacting says that the DME is loosing timing data from the CPS, or the DME is loosing power/ground.

The Cylinder ID signal from the #6 plug wire is used by the DME to time bank sequential injector operation. The engine will run okay without that signal, but performance and fuel economy will diminish.

[510man]

Where does the DME ground and draw power? Checking those connections would seem the next logical step.

[SmokeE30]

The inductive pickup or cylinder ID wire on the #6 plug wire won't cause the car to run that poorly. The aftermarket cps might, as it heats up from use the resistance values may be changing and throwing things off.

[jlevie]

Where does the DME ground and draw power? Checking those connections would seem the next logical step.

DME power is via the small cable from the battery. That wire includes a fusible link which can get corroded and cause problems.

[510man]

Still at this problem. I replaced the CPS again since "new" doesn't mean "good". I used a BMW OEM CPS this time. Cleaned the connection and the DME connections with electronic cleaner. And...............I still have the problem.

Took it to a local shop that has E30s all over the parking lot, including a few racecars. They spent two hours on it and couldn't reproduce the problem. They pulled a code 33 using an OBD-I code reader, which is a general timing intervention code. They checked the fuel pressure, which is fine and all the ground and positive connections at the battery, DME, fusebox, etc. and tightened them to be sure. The main power relay is original. They recommended I change it. Being a vert, they said there's no fuse link in the trunk (?) so that's not an issue. The previous post refers to this fuse link. Does the vert have this fuse link or not? They said they could continue at an hourly rate but until the car exhibited the problem, they were making educated guesses like I was. I picked up the car with no progress made. I'm just poorer for the two hours labor.

I picked that car up Friday at 6:00pm. It is now Sunday at 4:30PM. Since then, the car had the problem three more times in about 40 miles of driving. What have noticed now that I'm paying closer attention?

The gauge bounce is now the MPG gauge and the temp gauge. The other gauges do not move. Tach is no longer bouncing. The third time it acted up, it through a CEL and continued to miss. The other two times it missed a split second and continued to run well. On the third incident, I was two miles from the shop so I limped to their parking lot with the CEL on with the intent of leaving the car and calling them Monday morning now that car exhibited the problem on restart. I literally pulled in a parking space in their lot and the CEL clears. Car runs fine. What?!!! However, driving the car that short distance I noticed some things I had not noticed before.

The second incident was while idling at a light. Car missed a split second and cut off. Started right back up. Driving after the third incident with the CEL on, the car will run fine up to about 1800 RPM. Then it misses like it's in some kind of rev limiter or valet mode. I was able to drive the car fine provided I stayed under 1800 RPM. The moment I exceeded 1800 RPM, rev limiter time. It coughed and missed terribly. MPG gauge jumps to 40 and the Temp gauge needle moves back and forth a 1/4" or so. The other gauges do not move.

Does the E30 have a rev limiter or some kind of "limp mode"? My Mercruiser boat engine has a limp mode and I have an Electromotive TEC II system on a Datsun that acts exactly like this when the rev limiter is enabled or it is in valet mode and the RPM threshold is exceeded. If so, what invokes it on an E30? A faulty knock sensor maybe?

Nice car but I'm ready to roll this thing off a cliff!

[603Racing]

C101 plug under the intake, check it for corrosion. If it's corroded clean it to see if that helps. If it helps, then replace it with a weather pack connector.

[510man]

I don't have a C101 under the manifold. The only one I have is above the intake manifold by the thermostat housing. I opened that up and it's clean in indie. I've never taken it apart to clean it. Is that what you're suggesting or do some models have a C101 under the manifold?

[603Racing]

My bad, C191 connector. Commonly referred to as the "bitch plug".

[jlevie]

The previous post refers to this fuse link. Does the vert have this fuse link or not?

Yes

On the third incident, I was two miles from the shop so I limped to their parking lot with the CEL on with the intent of leaving the car and calling them Monday morning now that car exhibited the problem on restart. I literally pulled in a parking space in their lot and the CEL clears.

The code associated with the CEL will remain in the DME until cleared or the DME reset by removing power.

The moment I exceeded 1800 RPM, rev limiter time. It coughed and missed terribly. MPG gauge jumps to 40 and the Temp gauge needle moves back and forth a 1/4" or so. The other gauges do not move.

The movement of the temperature gauge is interesting, but probably an artifact of the cut outs. The MPG gauge going to 40 indicates that the DME isn't producing firing pulses for the injectors. In conjunction with cut outs that is most commonly caused by a problem with the timing reference data. Though one usually sees the tachometer react. The DME does include a rev limiter that is triggered by the engine speed as reported by the timing reference data.

With what has been done my suspect would be the engine harness. The coax in it for the CPS data may be degraded. A problem there will cause exactly the problems described.

Oh yeah. If original, the engine harness on this car does not have C191.

[510man]

My bad, C191 connector. Commonly referred to as the "bitch plug".

I have this plug connector. I've never opened it up but some of the C191 pics I've seen using Google make me think I should check it. there is noting like this under my intake manifold.