OK, I'm stumped - no spark zxt swap

[DCZ]

Here's the scenario -

 

73 240z with 83 zxt swap, stock ECCS. I've been DD this setup for two years without a hitch.

 

Died on road - tach does not move with cranking, no spark at coil secondary. Coil primary ohmed good, had 12V at + and - terminals.

 

Figured it was the ignitor, swapped in spare Z31 unit, no change.

Figured maybe the spare was bad, swapped in a new chevy HEI, well grounded, properly wired per diagrams found here and elsewhere (Thanks Skittle). Fired right up! Test drove it for about 10 mins and it died again, towed home, AGAIN same symptoms. HEI module was not hot to touch. HEI tested good at auto parts store.

 

Swapped spare coil - no change.

 

Ran troubleshooting tests in the FSM for ECCS system (FI relay, grounds, sensors, etc.) using volt/ohm meter. All passed.

 

Measured ignitor trigger signal with oscilloscope (ECCS pin 5). NO PULSE. Measured crank angle sensor signal to ECCS (pins 8 and 17) PULSE PRESENT! Concluded that ECCS bad, bought another (thanks TooSlowToCare). NO START SAME SYMPTOMS!

 

Ohmed out wiring harness for all signals between the ECCS and the coil, also the high tension lead from the coil to distributor. All passed.

 

Battery has a good charge and voltage only drops to around 11 while cranking, otherwise at 12.5.

 

I don't know what else to check. I find it unlikely that both ECCS control units are bad, and/or all the ignitors are bad.

 

The only two things I find that are weird are:

 

1) With any of the ignitors and coils, I have not been able to brush the trigger lead to ground and get a spark from the high tension lead held 1/4 inch from the radiator. Key in "run" position, brushing trigger lead to ground. As a matter of fact, I tried brushing the coil "-" terminal to ground and got no spark. Am I doing something wrong?

 

2) The crank angle sensor pulse voltage is 4 volts peak to peak measured at the ECCS. Should it be more? I measured 12V going to the sensor with the cable unplugged.

 

Any ideas appreciated, except "go megasquirt" as I'm not in a position to do that at this time.

[JSM]

I have an 82 280zxt all stock. I was driving down the road and it died. Would just crank and nothing. Got it home and realized I had gotten water into the distributor cap. And antifreeze does not seem to dry up!

 

I only bring it up as the rest of your investigation seems very thorough.

[DCZ]

Funny how they can go from being so solid one minute to "dead" the next. This car has been completely reliable since I did the swap 2 years ago. Started on the first twist of the key, ran great, hot, cold, no matter. Now it seems like nothing is working.

 

Guess I have to keep hammering away and re-check everything.

[Kash]

Man it sounds like you've checked all the usual suspects, check and double check all your ground contacts make sure the coil strap and igniter pack is still properly grounded. Sounds like the ecu but the new one did not work. Good luck!

[DCZ]

Didn't get to the garage last night to work on it. Just read old posts till my eyes bled about "zxt no start" and "z31 ignitor" etc. etc. I'm going to go over all the wires and grounds like you said, physically removing, cleaning and reattaching each one and verifying contact. I've already done that for many of them but I'm going to be a little more systematic. I'll post back when I find the culprit.

[DCZ]

Here's a diagnosis idea:

 

I measure no pulse at the gating signal for the ignitor (power transistor) at the coil. I figure this is due to either a bad ECU or bad crank angle sensor. CAS is producing a 4 volt square wave but I don't know if that is enough to trigger the ECU, and I've tried two different ECUs. I have also tried brushing the gating wire of the ignitor to ground (test from the archives) but that produces no spark so I still don't know the source of the problem (CAS, ECU or ignitor).

 

I have access to a function generator. If I hook that up to produce a gating signal for the ignitor, I should be able to make the coil spark, right? If I can get a spark this way, I'll know that the ignitor is good. Then I could hook the function generator to give a signal to the ECU to mimic the CAS, and see if I get a spark. That would test the ECU and isolate the fault to the CAS.

 

The only thing I'm not sure of would be the amplitude of the signal. I would guess that 0 to 5V square wave at about 100Hz would work for the ignitor (100Hz = 6000 cycles/min which would be like the engine running at 1000 rpm) and about 17 Hz for the crank angle sensor assuming one pulse per revolution (17 Hz = 1000 rpm). 50% duty cycle hopefully wouldn't harm anything.

 

Anyone know what the amplitude and/or duty cycles of these signals would be? Anyone know if I'm likely to destroy something by doing this? STOP ME BEFORE I BURN MY CAR UP!!