3.2 Carrera with a poltergeist

[GH85Carrera]

I tested every component I could and I could not find any strange readings. I am throwing in the towel on that step and I sent my DME in to be repaired. Bench test and fix whatever is wrong. If it comes back as OK, I have a known good start point.

I will post the final solution.

[brianlay]

good decision...let us know what happens

[scarceller]

Could very well be the DME coil drive circuit. I fix this type of issue often in these DMEs. Your best bet would have been to locate another 84-89 car and test your DME in that test car.

Also, an ICV failure would not usually result in a no start. It could cause a high idle or a stall condition after she catches. Your issue sounds like it may be no spark or weak spark.

[GH85Carrera]

I sent my DME in to Steve Wong be checked and repaired. His reply was:

I was going to let you know all is fixed and finished testing and full road testing out yesterday and runs perfect now. You had a failed ignition circuit and part of your problem was due to an out of spec capacitor that just ages out with heat and time, so that has been replaced with a new Japanese made Nichicon high temp capacitor which is the highest grade and temp rated capacitor available. You should be good now. I might suggest to ohm and check the resistance of your coil primary and secondary to make sure they are in spec, just in case so it’s not a contributor or stressing out the ignition circuit in the DME.

--------------------------------------

So soon I should have a running 911 again!

I will post a final wrap up with the final results after driving it for a while.

[DanielDudley]

Nice.

[AHudson]

Hope that fixed your problem, as mine is experiencing some similar sounding issues.

[mysocal911]

This was most likely your problem:



That is one of three connections for the ignition driver on the main circuit board that results
in about 90% of all 911 3.2 DME ECM failures. If one is having an intermediate running or a no-start,
most likely that's your problem. With a simple resoldering effort, it can save you $400 to $500
by avoiding a so-called "Rebuilt DME".

[GH85Carrera]

Well (insert your own long string of profanities) it still is haunted.

I found a used OEM Bosch black coil and put that in. Since I still can't find a reliable new coil for a 3.2 Carrera I went with a used coil.

I happily put in the DME with the fresh repair and full bench test from Steve Wong and went for the test drive. It ran like my 911 is supposed to while warming up. I was smiling and figured my 911 was exorcised of the ghost. It got to full temp and I was enjoying my car again. Driving along at 50 MPH a couple miles from my house it was like someone pulled the coil wire. The tech was still reading but there was no go pedal. I started my right foot madly pumping away and it would relight for a few seconds.

It would die often but fire up and go like complete normal. At wide open it did seem to run just fine. I rolled a couple of stop signs (it is a very rural area and there were no other cars in sight) and got the two miles home.

My garage in on the side of my house and it is easier to back into the bay I use for maintenance. As it sat there for the second needed to stop and get into reverse it idled fine but as I went to back up it died. It started up and died a few times before I backed into the garage. By the time I finished cussing and got out the test equipment it had cooled enough to have no issues to find. It is like chasing a ghost.

I am ready to throw some more parts at it. New flywheel sensors and new head temp sensors. I already have a spare fuel pump I take with me on my long road trips. I may throw that at it as well.

[gomezoneill]

I'd bet it's the reference/speed sensor they go bad when they get hot. Get the BMW one they're way cheaper.

[GH85Carrera]

Originally Posted by gomezoneill

I'd bet it's the reference/speed sensor they go bad when they get hot. Get the BMW one they're way cheaper.

Those are already ordered and on the way. I will do them this weekend.

[Omegaman]

Crank sensor can fail once the engine heat has soaked into them......id change them. I've had a few cars that have done exactly what your car is doing...both were caused by failing crank sensors.

[scarceller]

Do inspect the ref/speed sensors, if those harnesses are cracked and brittle replace the sensors.

I also suggest you jumper power from the rear fuse block directly to the coil's '+' post and see if that helps. I've seen issues with ignition switches cause intermittent issues. If you apply power directly to the coil's '+' post you will also power the DME relay's activation pin #86 and insure that the relay is ALWAYS active, even if the key is not in the ignition switch. It's a simple easy test.

The other thing to observe during the failure is if the DME relay is chattering? meaning is the relay clicking on/off when the problem occurs?

In summary: inspect the speed/ref sensors and then also jumper 12v directly to the coil's '+' post and see if that helps. Remember to remove the jumper when not running the motor, else you will kill the battery! You can use a basic alligator clip test lead, be sure the wire is no smaller than 14gauge.

[scarceller]

Also, could be a fuel issue. You may need to plumb up a fuel pressure gauge and test FP and hopefully observe pressure during the failure. I've also seen a problematic injector cause NO fuel in all 6 cyls. If 1 injector shorts out it takes all 6 out! an intermittent shorted injector can kill fuel completely! I've seen this a couple of times. You can ohm out each injector they should all read the same (about 2-3 ohms), if one is way off you have an issue in that injector.

[gomezoneill]

Originally Posted by GH85Carrera

Those are already ordered and on the way. I will do them this weekend.

I'm sure that will fix your issue.

[mysocal911]

Originally Posted by GH85Carrera

Those are already ordered and on the way. I will do them this weekend.

Sounds like the DME ECM wasn't the problem and it didn't really need any "rebuilding".
It's unfortunate that you didn't have access to another DME or your "rebuilder" didn't
supply you another rebuilt unit to confirm the problem before rebuilding your unit.
Maybe your "rebuilder" will provide some form of a refund.

[scarceller]

Having the DME looked over is not a bad idea as they are 30+ years old. I've seen plenty of DMEs with cracked solder joints and still working fine but finding these issues before you have a problem is a good idea.

I'm not sure what 'rebuilt' DME is? I doubt anyone is rebuilding these units, at most they are looking for typical problem areas and doing repairs. My definition of rebuilt would be replacing all capacitors and any other aged out components or better yet having new boards built. I think calling the service offered by most a rebuilt unit is simply dishonest if they are not replacing all the aged components.

I do not rebuild the DMEs, I inspect, test and repair as needed. I'm very upfront if I help someone with a DME. I also from time to time upgrade older early DMEs to the 1987-89 28pin chip from the older 24pin.

[mysocal911]

Originally Posted by scarceller

Having the DME looked over is not a bad idea as they are 30+ years old. I've seen plenty of DMEs with cracked solder joints and still working fine but finding these issues before you have a problem is a good idea.

It's very rare that the DME has cracked solder joints other than the area pictured up-thread
by the ignition driver. Many Pelicanites could easily check that area of the circuit board
when DME problems are suspected without wasting money by sending the DME out for
a needless "rebuilding".

[stephen1070]

Originally Posted by mysocal911

Sounds like the DME ECM wasn't the problem and it didn't really need any "rebuilding".
It's unfortunate that you didn't have access to another DME or your "rebuilder" didn't
supply you another rebuilt unit to confirm the problem before rebuilding your unit.
Maybe your "rebuilder" will provide some form of a refund.

I read above that the rebuilder found a failed ignition circuit and an out of spec capacitor in the DME. That was replaced with a new Japanese made Nichicon high temp capacitor.

The rebuilder also suggested checking the resistance of the coil primary and secondary to make sure they are in spec so it’s not stressing out the ignition circuit in the DME.

The rebuilder does not owe the rebuildee a refund.

[AHudson]

Originally Posted by mysocal911

Sounds like the DME ECM wasn't the problem and it didn't really need any "rebuilding".
It's unfortunate that you didn't have access to another DME or your "rebuilder" didn't
supply you another rebuilt unit to confirm the problem before rebuilding your unit.
Maybe your "rebuilder" will provide some form of a refund.

Can't figure out why the "rebuilder" should "refund". He/they were hired to do a job after lengthy diagnostic and they performed that job, thus due to be paid. It seems you may have a "problem" with these solutions in general.

Back to OP: Though the DME was not the prime culprit, at least you now know that you have a nicely cleaned up and corrected unit to last you many more years. Value comes with confidence in that.

Hoping the new round of parts brings you an answer to this! In a similar but not same boat of going down the diagnostics trail. I too will be getting my DME refreshed (or whatever the acceptable term is) and whether or not this resolves it, it's time for it.

Keep us updated, best to you.

[ischmitz]

One "honest" way to segment a DME from other components is to offer a loaner DME for diagnostics. I have done this several times both to prove the DME was NOT the issue and to show a good DME is running a car.

The debate of rebuilding vs. refresh is pointless when it comes to electronics for the most part. Most components have a bell curve for reliability with high infant mortalities and then a very long if not infinite life time. Proactive replacement would introduce more issues. In addition to this many components are no longer manufactured and I have to work from NOS inventory. I offer a warranty instead and this is a value in itself.

When it comes to soldering there is a difference between re-heating old solder with a 100W iron meant to repair gutters and properly removing old solder and soldering with correct tools and materials. I have seen many sloppy jobs and that's a poor excuse for a rebuilt.....

In 9 out of 10 cases one can spot solder issues with a good stereo microscope but there are cases where even that analysis is not 100%. Especially when it comes to highly intermittent issues. I am trying to be very honest and upfront with customers and have sent DME back w/o repair if there was nothing apparent. This helps to keep the search on the right track. Nothing worse than sending back a unit with work done and it fixed NOTHING.

Replacing capacitors is not always warranted. Failures of electrolytic capacitors in modern consumer electronics are due to their use in buck-boost converters. This application is stressing an electrolytic cap often way beyond its intended use case if its ESR is not suitable for this application. No so in the DME - here caps are used primarily to decouple supplies and take the occasional glitch. I yet have to see one that failed in a DME.

Most repairs I do are hard-down units where a semiconductor device has failed and the fault is readily related to the failed component. In addition if someone decides to have a unit gone through and re-soldered I am very upfront about the fact that this won't make the car run smoother or give you the extra 10HP everyone is looking for. All it doses is giving you a little bit of "piece of mind" and a warranty. Plus I work with the individual until the case is solved.

Ingo