Challenging problem! As I have been following the thread, it occurred to me that maybe the DME failed at the moment you turned off the engine after that 30-second test run. You mention there are a lot of loose wires in the engine area. Is it possible that there is some mis-connection or missing connection that allowed the battery/system voltage to become un-regulated ? Perhaps it pushed up too high and damaged the DME or at the instant of shut-down a voltage spike occurred that took it out? Seems like the only reason you have no spark would be bad crank sensor or a dead DME.

Part of the problem is not knowing what problems would cause the DME to not send a pulse trigger signal to the coil.

If something were wrong with injectors, would that cause the DME to not send the pulse? Which other engine sensors would cause the DME to not fire the coil if something were wrong with them?

Mysteries.... I am going to try and see if I can find someone local with a 3.2 that I could swap my DME into and see if it runs.

Well if you can get a working DME and test it in you car it would be good but the odds that you will find someone to lend you one are slim.

Bad injectors would not cause a non start problem.

While some of the items you tested are ok statically, they might not be under power. The fact that it start up once it might be a clue that something fried under power.

Before you take it to a $100+/hour shop, I would get new; DME Relay, Coil, Coil wire, Distributor Cap & Rotor. If it still a no start, well you tune up is done and you get spare part for the next time you'll need to isolate a fault.

Put an '85 3.2 in my '73 911 then ran it for several months. Pulled it out and put it in my son's '71 and no start. Bought and replaced all the sensors, no start.
Started checking all the ground wires on the wiring harness and found one broken. Old copper is brittle. Fixed it and engine fired immediately and hasn't failed to start for 8 years running.
If all your feed wires are OK check your ground wires, they are there for a purpose.
Good luck, I was ready to blame Gremlins.

check this parallel thread - some excellent links in embedded answers

http://forums.pelicanparts.com/porsche-911-technical-forum/767428-84-carrera-no-start-questions.html

http://forums.pelicanparts.com/porsche-911-technical-forum/738792-3-2-starts-dies-immediately-all-tests-good.html
.

Ok... some progress. I called up an old racing buddy who has a 3.2 track car. Drove down to his house with my DME. Plugged it into his car. Nothing - would not start. Plugged his DME back in - fired right up.

Bad DME.

So - Where is the best place to source a new one or ship mine out for repairs??? I am in Atlanta.

Since you already open the DME and look for cold solder joint problem, one more swap before sending it for repair... DME chip.

That should be easier to find as everybody that upgraded to a performance one (like SW one) still has their original.

eBay - several rebuilders advertise there - about $750 with your core

I haven't dealt with either of these companies, but here are two Pelicans that contribute information to the Forum.

Systems Consulting - Products

911Chips.com - Porsche Motronic ECU / DME Repair

No pressure - but the Petit Le Mans countdown has begun - Petit Le Mans

Mike

Right here at PP, $450 and 2 years warranty; Pelican Parts - Product Information: DME-REBUILD

TOTAL NIGHTMARE GRRRRRRRRRRRRRRRRRRRRRRRRRRRR

So I shipped my DME off for repair, and a fellow P-Car nut who had an extra offered to let me borrow his to get the car running and driving until mine came back. So installed the DME he lent me and the car fired right up and ran great.

I take it down off the jack stands, get the drivers seat bolted back in and go for a test drive. Car runs great! Awesome!

I pull back into the driveway, shut her down and come back out about 30 minutes later to go get some gas.

No start. Exactly like before.

I am going to again remove the DME and see about heading to my buddies house to test it in his car again, but it appears that there is something wrong in this car that intermittently causes it to fry the DME.

So every time you want to test a possible fix - if you have it wrong, it costs you $450.

What could possibly cause damage to the DME?

Any ideas or suggestions are greatly appreciated.


Terry

Not sure if anyone is following this thread any more or not.... it does seem a lost cause.

I took this second DME down to my friends house to give it a try in his 3.2. Surprisingly enough - his car fired right up with it. It did not seem to idle all that well... but it certainly fired up. In my car it does not start and gives zero indication of starting.

Utterly baffled at this point. Steve Wong has my old DME and is digging into it, hopefully what he finds wrong can help point me toward whatever is wrong.

A bad alternator putting out too much power will fry the brain..
Bruce

I know you check it, but did you ever change the DME relay?

3.2 test stand
 

Terry- If you don't get this solved and cold weather sets in-consider building a test stand and setting up your engine DME coil fuel pump etc on it. I know that is a lot of backtracking-but it may be worth it. Then you will have full access to check output of all the sensors, and to the DME pins. Here is a photo of a test stand I made out of angle iron for my 3.2-it allowed me to fix a couple of issues before the install. I used a starter ring adaptor, and made a small fuel tank.http://forums.pelicanparts.com/uploa...1378467433.jpg

I would start by confirming the wiring for the ECM.

Disconnect battery and ECM.

Take a ohm meter and connect 1 lead to each pin of the 35 pin connector of the ECM plug. (Pin 1 is the coil)

Disconnect the wire from the device at the other end (coil) and put the other lead at the other end of the harness. (disconnect coil wire and connect to meter)

Make sure that you have ~zero ohms through the harness, and that all the other 34 pins on the ECM connector have infinite ohms to confirm no shorts through harness.

Move on to pin 2. (idle speed switch).

Mike

Verify the alternator, regulator, battery are putting out correct voltage. I would say to make sure every ground is clean and solidly connected. Transmission ground strap, terminations on intake runner, rear electrical console ground strap, alternator ground strap, all the pins in the 14 pin connectors and the connector to the injectors. Check with 12v to injectors to verify each one clicks open and none are shorted or sticking.

I can think of something that may cause these symptoms - if the reference or speed sensors are too far from the flywheel, they will give a weak signal. I could see this weak signal 'sometimes' being read and sometimes not, and I could see some DME's being able to trigger on the signal, and some not. Maybe the ref sensor holder is loose? Just something to check, I haven't reviewed all the posts to see if this has been done. I always try to get sensors like this as close as possible to their trigger wheels.

HTH,
Chuck.H
'89 TurboLookTarga, 362k miles

Another DME relay nightmare! Almost like post 54

Steve Wong has my original DME for repair.... says it is toast, the primary chip is fried, caused apparently by attempts to make the signal from the DME work with a '73 model tach. The tach has never worked since I owned the car but I know the prior owner attempted to wire the stock '73 tach to the DME tach wiring without success.

None of this makes any sense at all to me. The car DID NOT stop running... it refused to start again after being shut down. The problem has never been the car dying while running but always refusing to start - even if it did start only moments before. If something was fried in the DME, would the car not have stopped running the instant the frying occurred???

I tried replacing the relay - no change, still will not start.

I have tested and checked pretty much everything there is to test or check on the car. It simply will not start.

At this point, I am seeking some Porsche technical GOD to try and take the car too for repair - OR - planning to tear it all apart again and launch into the process of replacing this Motronic CRAP with carbs and a sane ignition system.

Where are you in Cumming?

Interesting comment on the tachometer. My conversion ran fine using the stock 1973 tachometer.
Maybe if the original tach already had problems, but you can try buying a good used Carrera tachometer.
1973 Porsche 911 - YouTube

Sully,
NICE car. it looks like you could eat off of that motor.


Tspringer,
Too bad the ECM is fried . Just remember that before you throw away the "Motronic crap", it still has value - even the non-functional ECM. Since your tach didn't work, did you get a chance to check pins 11 and 21 of your ECM connector to make sure you don't have a problem from a PO?

Mike

I went through and started testing everything again - a back to basics full "reset" on the testing process.

I think the issue is related to the reference sensor. In my notes from testing before, I have a check mark next to this one indicating it tested to spec. But when I tested today, both at the DME plug (pins 25 and 26) and at the engine harness plug (bottom of the 3 plug in connectors) I got nothing. No continuity and 0 Ohms - an open circuit. Spec appears to be .6 - 1.6 k Ohms.

If this is correct I have a bad reference sensor. Perhaps a bad internal wire connection such that sometimes it makes a connection and other times not?

The gap seems correct, but I will reset it using the paper method after I remove the sensor itself.

Anyone have suggestions for a more definitive test (if possible) to tell if this reference sensor is no good?

Tspringer,
It sounds like you're making progress - that's good news.

I haven't tried this, but if I wanted to test a reference sensor, would do something like this -

http://forums.pelicanparts.com/uploa...1379223601.jpg

I believe if you put a piece of metal in front of the sensor, the voltage at the meter will change. Maybe someone can verify this.

Mike

I did a LOAD of reading on the speed and reference sensors when I was going through my no start problem. It sounds like the only real test is using an oscilloscope. I checked my NEW sensors with a Ohm meter and would have said they were bad based on my reading of the subject. IF you decide to go with new sensors, Pelican sells a sensor for BMW's that is half the cost and works fine. I'll try to find the part number if you go that route.

My problem was part the fact that my wires were crossed so the speed was going to the reference and the reference to the speed sensor. The other problem was the cursed DME relay that I would swear to this day was good, would not let the engine start. Make sure your DME relay is good. I can't stress that enough after spending a few days searching for everything else and it ends up being the relay. I have read where the DME relay can be a hit or miss thing meaning one day it works and the next it doesn't.

I hooked up the circuit from my previous post, using a good used sensor. The sensor responded to metal moving by the sensor with a momentary increase in voltage measured on a voltmeter.

When I was troubleshooting my Motronic issues, I just bought 2 new sensors, because they were 25 years old, and provide the timing for the engine. If you come up with an oscilloscope, and get to the signals (PITA), Here's what it should look like -

http://forums.pelicanparts.com/uploa...1379292867.jpg

Mike

Interesting stuff Mike...

I did the test for the reference sensor with everything hooked up first - tested pins 25 - 26 at the DME plug - got nothing. Same test for pins 27 and 8 for the speed sensor and I get just over 1K ohms, exactly as the testing chart Sully posted said I should.

I removed the reference sensor over the weekend (boy that was fun) and tested it. On the bench I got the same results - no continuity and 0 ohms across any connection.

So I decided to cut the wire and test it. There are 3 wires in there.... yellow, black and a bare metal wire wrapped around both with the entire grouping wrapped in aluminum foil and plastic. I was not expecting that.

I tried testing at the wires and got nothing... but honestly not sure how to test the 3 wires.

Mike - in the diagram you posted, which 2 of the 3 wires were you using?

My best guess is that the reference sensor is bad, but other than the Ohms test based on the troubleshooting chart that Sully posted I have no means of really knowing.

Whats the BMW part number?

I used 2 of these on an 88 back in January, they were a direct replacement for the originals -

Pelican Parts - Product Information: 12-14-1-708-619-M14

Of the 3 wires on the sensor, I used the 2 that had ~1k ohm across them. I used 5 volts because that's what powers the sensors in the ECM.

Mike

Thanks Mike. I ordered a new sensor online yesterday after searching google on the part number listed before. It was $63 including shipping.

My bench testing of the old sensor shows 0 ohms across the two wires (ignoring the shielding wire). I tried waving a hunk of metal in front of it and sticking it to my giant steel vise - nothing magnetic made any difference to the ohms reading. Given that the speed sensor when tested does show the correct ohms reading, my assumption is that this sensor is bad.

Steve Wong called me back about my DME - it is totally toast, cannot be repaired. He said that the tach circuit had been plugged into an older style tach (the PO did this I know...) and that had fried the DME. He has shipped me a totally new, updated unit. When it arrives and the new sensor arrives, I am hopeful to have this all fixed.

It is still all very strange. The reference sensor tests bad - yet the car did start before - sometimes - with 2 different DME's.

Terry,
I am still curious about the Steve Wong thing that using the old Tach burned out the DME. I used the original '73 tach, and 2 others that converted also used the original tachometers, '73 and '72, without any problems burning out the DME. The one using the '72 tachometer says his reads a little high, but no other issues.
I hope the new sensor gets you back on the road.

Steve Wong told me that hooking an early model tach up to the black/purple tach wire from the DME OR to the solid black shift light wire from the DME could either way fry the DME. He believes that is what happened to mine. He says the black shift light wire should not be used to try and get an rpm signal or bad things can result. He indicated to me that this was a common problem.

I have no idea about this. Neither wire has been hooked up to my tach since I owned the car, but I do know the prior owner tried to make the original '73 tach work and could not get it to do so.

Other that what Steve said to me... I have no clue. I have the "new" DME back from Steve and only need the new sensor to put it together and hopefully have it fixed.

Fixed!

I installed the new DME from Steve Wong and the new reference sensor. Car fired right up and runs great. It runs much smoother than before.

I have driven the car around twice, started and stopped it over a dozen times and it now appears to have no issues.

My best guess is that the old reference sensor was bad but that it still worked - barely - sometimes. Also, the old DME was a mess and not functioning properly and somehow in the process of trying to diagnose problems it finally gave up the ghost. I believe there were thus multiple problems at the same time that caused the diagnostic process to be confusing.

Anyhow, she is running great now!

My '72 tach works fine with my '84 DME. I didn't use the shift light wire.

Roger

I can see how miss hooking the Black shift light may cause a problem with the DME, but hooking up the Bk/P does not seem to.
Anyway, it's good to see you have the car running. Now you get to enjoy it.