No Ignition Spark in my 911

[Steve N L]

I have a 1987 911 that lost ignition spark last October and I have been slogging my way thru the Motronics system since then. I wouldn't bore you with all of the details, but I seem to have ruled out almost everything. The coil is good, as is the ECU and the fuel pump/ECU relay. I replaced the crankshaft sensors, although they appeared to be OK. I have checked all of the grounds and connectors I could find. I have current to the positive terminal of the coil when the ignition is on.

From the coupler on the ECU, I have checked all of the grounds and they are good. I have checked continuity on the green wire from the ECU to the negative terminal of the coil and it is good. From the ECU coupler, the resistance for the reference sensor (pins 8 & 27) are 1,000 ohms. The resistance for the speed sensor ( pins 25 & 26) are the same. This seems to indicate that the connections to my crankshaft sensors are good.

There is one thing that is clearly wrong. When I cranked over the car with an oscilloscope attached to the negative terminal of the coil, I found that there is no pulse going from the ECU to the coil. A multi-tester shows that current from the ECU is 12.5 V at the beginning, but drops to 10.5 V. It’s a flatline. Nothing is happening to collapse the energy field in the coil.

One last thing….While doing diagnostics on the car, it inadvertently started and ran nicely for about 20 minutes. Then it died and lost spark again. It has not started since.

Does anyone have suggestions as to where I should go next with this? I'm running out of cards to play. Why would I get no pulse from the ECU?

[Steve@Rennsport]

Steve,

The coil driver is incorporated into the ECU so unless you have a form of "connector-itis", your ECU may need repair.

[Chuck.H]

Steve -

Open your ECU (also called a DME by Bosch) and look where the stand-off-the-board coil transistors (on the heat sinks) are soldered to the circuit board - these like to develop cracks in the solder, resulting in a on-off no spark condition, exactly like you describe. You'll have to remove the cover and the bottom plate to see them, but this is simple disassembly.

This happened to mine at 200k miles, I reflowed them (touch with a hot soldering iron) and it's been great since.

Hope this helps,
Chuck.H
'89 TurboLookTarga 332k miles

[Steve N L]

The problem may be my ECU, but when I plugged it into another car, the car started right up and ran. When I plugged it back into my car, the car still would not start. That seems to imply that my ECU is good.

Here is one other symptom. I hooked up an oscilloscope to the new crank shaft sensors that I installed after this problem began. When the car is turned over, the pulse for the speed sensor begins as a castelated pattern, but deteriorates to waves then tp a flat line. The reference sensor pattern is a flat line with an occasionsl blip. I think there is a problem with the pattern from my speed sensor.

[Lorenfb]

Check here for waveforms of the key signals; Waveforms

[Steve N L]

The problem turned out to be my ECU. If the damn thing had not worked when I plugged it into another car, I would not have spent many hours looking through every other aspect of the ignition side of my Motronics system. All of the diagnostics continued to lead back to the ECU, which I continued to believe was good. Another learning experience.

[Justin@Athens]

Did you take it apart and examine the soldering to see if that was the root of the failure?

Yet another reason these should at least be isolated from the floor with rubber grommets or some other material.