Stranded! DME Ignition Guru Help needed

[Thrasher]

I got up at 7 am on Friday, drove to meet a buddy for 9 holes of golf. Car started and ran great on the way there. I was running late, so I shut it off and grabbed my stuff, played 9 holes returned to car. It was pretty warm outside (90 F), but that shouldn't be an issue.

My buddy had never seen the car, so he sat in the driver's seat and was loving it. He asked me to start it, so I put the key in the ignition and cranked... and cranked... and cranked. nothing. I suspected the DME relay (which is only 6 months old), but I definitely could smell raw gas from the tailpipe. I waited 1/2 hour in case it was flooded. Same thing. Called the flatbed and 4 hours later the car was in my garage.

Started diagnosis Fri night and determined no spark. Ohm-meter tested cap and rotor; dressed contacts, but they all looked ok. I took the primary (coil) lead and put a drill bit in the socket to act as a conductor, had my wife crank the starter with the drill bit 1/8" from ground and nothing. But, when she turned it off and back to "run" a big, fat blue spark arced the distance. Cranking, nothing. Turn it back off and on, another big fat blue spark, so I'm thinking the coil is good. I verified 12.6V at coil per Bentley.

So, I'm now thinking 3 suspects: ignition switch, flywheel position sensor and, of course, DME.

1. If I'm getting 12.6V @ coil with ignition switch in "run", can I rule out the ignition switch?

2. There are VOM resistance tests for the sensors, but the harness is really tough to get at, so I'll have to remove my heater ducting and then remove the harnesses. My initial testing on the upper one sure looked like infinite resistance, but I had to use a mirror to see if I was making contact, so I can't rely on that as accurate.

3. The DME: I replaced my Autothority chip with the SW chip just about 200 miles ago, but with no problems since. The change also required me to re-jumper the DME circuit board to use 4k. Maybe my soldering job has failed?

After re-reading more Bentley, I found two interesting passages that say the crankshaft reference sensor is only used at starting time, to give an initial position. Without that sensor the car will not start. But, would the car have continued to run fine if that sensor had died on Friday morning during my trip to the golf course? It sounds like that may be the case. So, possibly either the sensor went bad or the slug that it picks up a signal from fell out (unlikely).





Where should I concentrate my efforts? It sounds like you can check the Reference sensor's operation using an oscilloscope while cranking the starter - can I use a multimeter for this test to look for the "pulse" when the sensor "fires"?

[Thrasher]

Here is a pic of the Bentley troubleshooting matrix. Note the test for the reference sensor - how would a sensor firing once/rev create a sine wave?

[DonDavis]

Pull a plug wire, put a spare spark plug in it and place the plug body on a metal "ground" location. Have the wife turn crank. If you have spark, your reference sensor is fine. During cranking the ref sensor looks for the "slug' and if its there, the ECU allows spark. My brain is tickling a memory of the Cylinder Head Temp sensor failing and causing your situation. I can't remember fully. Check your Bentley for more info on that.

I know you said there was a fuel smell but you should check the fuel pump voltage AT the pump. You need to verify its running.

Check for loose connectors or wires everywhere.

If you have spark and the pump is running, next check for fuel injector pulse with O-scope.

Good luck.

[Thrasher]

Don, did the spark test - no go. that's when I proceded to the main lead from the coil; it only sparks when key switched to "on", but never gets a signal to fire when cranking.

SInce I know there's no spark and I feel comfortable that fuel is being delivered I have concentrated on getting a spark.

Can anyone verify that a faulty CHT would prevent spark? Bentley did not even mention looking at the CHT sensor on ignition issues.

I will also look for the slug on the flywheel, but I'd put the money on the electrical components as opposed to that falling out randomly after 22 years - but you never know!

Can I access/view the flywheel sensors w/o an engine drop? I assume they're under the car? Bentley is very thin on info about their location/access.
EDIT - nevermind - I found this info in the archives

I am 99% sure it's the reference (TDC) sensor or DME (or assoctiated wiring) at this point.

[Steve W]

Since you seemed unsure of the soldering job you did on the DME jumper, I'd go over it with a soldering iron once over to make sure it isn't something as simple as that, and check that there isn't any trace solder jumpering across where it shouldn't be.

[DonDavis]

Originally posted by Thrasher
SInce I know there's no spark and I feel comfortable that fuel is being delivered I have concentrated on getting a spark.

If you are not getting spark, the fuel pump should not be running. Once all conditions are met, (i.e. ref sensor signal present, voltage to DME relay correct) the ref signal is created by the slug passing the sensor, the fuel pump runs and spark is allowed. If you don't have spark but fuel pump is running, I would say the ref sensor is ok and I would swap in a known good coil. Or put your coil on another 3.2 car.

[ZOA NOM]

Yes, the DME requires an input from the CHT to allow the car to start. The chart above shows the temp sensor as a suspect under the "will not start or starts hard" column.

[aleng888]

you should try remove you rotor cap and whip in dry. most like water got into it.

[xlr8]

I had a very similar no-start issue with my '88, it was cracked solder joints in the DME brain. I needed sunlight and magnification to see the cracks.





Brain surgery should be a last resort. I was able to isolate this problem with a known good DME brain. These cracked joints made intermittent contact. The car would start and run when parked in the sun, but not in the shade on a cool day. I have had no problems since re-solder 18 months ago.

[movin]

Had similar issue and it turned out to be the DME unit had faulty connections somewhere on the circuit board. A brand new DME solved the problem.
On another note what are you talking about having to solder something to accomodate a SW chip???? "re-jumper the DME circuit board to use 4k"??? I never heard of this for any Carerra DME, weird. Please explain.

[Jeff]

I had a the same thing happen when a metal plate on the bottom of my seat cut a wire going to my factory alarm. It shorted the alarm module and in turn armed the alarm. I had a no spark issue (not sure if it sparked in the start position like yours) and went as far as replacing the coil before I bypassed the alarm. The reason I bring this up is you said your friend sat in the driver's seat when you tried to start it. Is he bigger than you? ...

[Rot 911]

Craig, not sure what your problem is, but if you need to swap out the CHT or speed sensors let me know. I am pretty sure I still have my old ones from when I rebuilt my 3.2. They were working when I removed them.

[Thrasher]

Thanks for all the great responses.

Last night I pulled out all my heating ductwork and got a better angle on those sensor connections so I could be sure my multimeter probes were on the contacts. I confirmed that my Reference sensor is DEAD. It reads infinite resistance across any of the three terminal connectors. The speed sensor ohm-tests exactly as Bentley describes.

Strangely, the wiring at the sensors looks to be completely intact. It must've failed internally or the wire(s) just cracked from long-term exposure to heat, etc.

Steve W. - it was working great since I got the chip from you in January, so I'm hoping it didn't suddenly go tits up on me.

AllenG888 car has not seen water (even washing) in months. Drove thru no puddles,etc; there is no spark

movin: early DME's only utilized only 2k of the available 4k of memory; Steve "enhances" the chip program with additional data that improves idle, etc, but requires you to reposition one jumper on the circuit board. It's an easy solder job and it has worked great since I upgraded my chip in January.

CUL8R: yes, he would've moved the seat and outweighs me - I had the same thought, but I have removed the old Ungo Box long ago and there are no other wires down there. Good thought, though!

XLR8: I saw your thread about the DME - if I get to that point it will be very helpful.

Zoanas: I think my CHT sensor is ok, and would NOT prevent spark, but may prevent the car from starting. All my symptoms point directly at the Reference sensor, which just happens to not test as it should!


Kurt: I will email you about those sensors.

Once I pull the sensors and replace I will report back with the results.

[DonDavis]

You can swap the ref and speed sensors and verifiy your findings. It should start but run crappy. Or it should at least try to start. Did you remove the rear wheel to get at the sensors? I found that was the easiest way to access them.

BTW, if you do the swap, don't forget to swap the connectors!

[Thrasher]

That's a thought I had not considered. Since my ref sensor is completely dead, will it actually even run using the dead sensor in the "speed" position? It would prove my theory about the lack of spark, though. I could swap them and try cranking and look for spark.

But, then I will be messing up the clearance and I read that using the old sensor was the easiest way to set the clearance on the new one if you're not dropping the engine and can easily set it.

[DonDavis]

If you have the rear wheel off the access is not too bad. Setting the clearance is a little bit of a pain but very do-able. If you want, just back out the allens and that won't mess with your settings. The sensors bottom out in the bracket. FYI, the top hex bolt hole is round and the bottom is oval. The bracket pivots and the setting is made at the speed sensor.

[Thrasher]

Right - I pulled the LR wheel last night and could see the sensors. I have B&B headers and the heatbox tub is pretty much right in the way. I removed the flexible tubing that attaches to it, but it's gonna be tight!

That's good to know about the settings and holes. So, if I assume my current sensors are set properly, and the parts are interchangeable (same length) I should just have to drop the new ones in pre-set the same? i.e. w/o having to touch the bracket screws?

[ZOA NOM]

Originally posted by Thrasher
I confirmed that my Reference sensor is DEAD. It reads infinite resistance across any of the three terminal connectors. The speed sensor ohm-tests exactly as Bentley describes.

Zoanas: I think my CHT sensor is ok, and would NOT prevent spark, but may prevent the car from starting. All my symptoms point directly at the Reference sensor, which just happens to not test as it should!

The chart above says to check the reference sensor with an oscilloscope for a sine wave, did you do that?

Also, you can place a resistor across the connector for the CHT to bypass it and confirm it is not the problem. I think 270 ohms is the value required. If the CHT is bad, it could cause what looks like a no spark condition, because the DME will not give the ignition system the command to fire.

[Thrasher]

No, but if there's no continuity, there ain't gonna be no sine pulse to measure. The reference sensor is toast. If KurtV's known-good used sensor still doesn't fix my issue, I will get access to an O-scope and test it dynamically, but I don't see the point when it fails the static (ohm) test.

On the CHT: I saw the resistor "fakeout" method in other threads. I will run an ohm test on the CHT while I've got it all out, but it is not the original single-wire unit and looks outwardly to be ok. It's worth checking though.

Thanks for your help.

Craig

[Thrasher]

For the archives:

Kurt V sent me his used, but good, flywheel sensors (Thanks Kurt!). My sensor looked pretty darn good and his... not so much. But, mine was dead and his ohm-tested ok. I wrapped his up in electrical tape and swapped out only the reference flywheel sensor (for TDC). The car fired right up.

Here's the way I diagnosed:

1. The car had started and ran perfectly before shutting down normally.
2. The car cranked strong but would not "catch" at all. Never even a single pop, but I smelled fresh fuel at the exhaust.
3. I pulled a plug - no spark. I pulled the main lead from the coil - no spark there. I had voltage at the coil AND when I turned the key to "run" from "off" with the main coil lead near ground, it flashed a fat blue spark. I put a drill bit in the coil lead (distributor end) and held it 1/4" from ground using a piece of wood, having my wife turn the key/run the starter briefly.

The key was learning that the reference sensor ONLY is in use while starting - that's why my car ran perfectly up until I shut it off and wouldn't even pop 2 hours later. The sensor died sometime between starting it at 8 am and shutting the car off at 8:30 am - and this never affected its running.

I am ordering a pair of new sensors today and will put them both in, retaining my speed sensor for a spare. Carrera owners should all be aware that these are basically a consumable part and look to these when no-start happens.

FWIW, mine looked basically new - always do the ohm test at the connector in the engine bay. I found it impossible to do with the heat ducting in the way. You'll need to pull the connector from the bracket by removing the metal 3-plug retainer. Of course, each connection has its own metal clippie thing, too.

Mine both had the yellow "DG" and "BG" stickers on them, indicating they *may* have been original, but the casing was so undamaged they almost certainly were replaced at some point.

I will document the replacement process with photos and write it up tech-article style.