KenMathisHD

ok, so, my cars been down for about a few weeks now (haven't had a lot of spare time recently) as I've been going through the process of troubleshooting the ignition system. It's a 94 Automatic with 137k miles

The problem I'm having is I'll be driving it as normal, but once the engine starts getting really hot, around 240 (gotta love these TX 90+ degree summer days), the engine will develop a severe stutter when giving it any amount of gas. Think of what would happen if the ignition system lost power for a moment, and then repeat that for a few seconds until the car dies. It doesn't always kill the car the first time it stutters, but eventually, the car dies. Once it dies it won't turn over until the car has cooled completely

I researched and looked around and got a hold of an ignition checklist off of AutoZones website
So I followed it and checked for spark at plugs while cranking, nothing
Checked spark at coil while cranking, nothing
Checked for voltage at the ICM connector on points A and D to ground, got 10+ volts both sides

At this point, I decided to replace the ICM since it was cheaper than the opti and figured maybe I'd get lucky, but no dice
While messing with the ICM, I popped the connector from the opti and noticed that the female connector for the pink&black wire was smushed, so I replaced that and checked again, but she just cranked and cranked and cranked
So I checked at ICM connector point b to ground while cranking, 0 volts

Next step is to check that low res pulse is being sent with a tech 1, but none of the local parts groups have a tech 1.
I'm thinking that if low res signal wasn't being sent/received that maybe the ECM would throw a code I could find with a regular scan tool or by using the paperclip trick? Or maybe with computer software and an aldl cable?
I'd like it not to be the Opti so I don't have to go through the headache of changing it, but it seems most likely from the data I'm getting.
What are you guys/gals two cents on this?
~KenMathis

Renfield

Not sure what a "Tech 1" is. I used a NOID light to check for an injector pulse. The Optispark failed in my '95 at 137K miles.

Symptoms were a little different, but also involved stalls at running temps. and hard restarts.

So she got a new Optispark, seals, hoses, fluid, knock sensors, plugs, wires and a coil.

Good luck!

billschroeder5842

I think you will be changing the Opti.

KenMathisHD

That's what I was thinking, but the problem seems different than all of the Opti fails I've been reading threads on, don't they usually start to miss during different ranges and slowly get worse?

MatthewMiller

The coil is "upstream" from the opti. If the coil is producing no spark when cranking, as the OP says, then the opti isn't the problem. Does the new ICM exhibit the same problem, it runs cold but then malfunctions when hot? Or does it malfunction even when cold? If the latter, you may have purchased a dead ICM. I got one like that. If you can find a place that can still test your ICM, have them test it for 8-10 cycles in a row to get it nice and hot. If it still shows good, then maybe the ICM isn't the problem.

KenMathisHD

When I swapped in the new ICM, the engine was hot and wouldn't start and the ICM was cold. Cooled down the old ICM and popped that back in there while the engine still wouldn't start and it didn't start up, only cranked like before, so if it was the ICM getting hot and clocking out, shouldn't cooling it have gotten the car to fire up?

KenMathisHD

Here are the diagrams and checklist I pulled off of AutoZone's website, probably should've attached these earlier
From looking at it, if B isn't producing voltage while the engine is cranking but not starting, would that mean that the ICM isn't getting a signal from the ECM to tell it to tell the coil to fire?

MatthewMiller

Yes, one would think so. As to your question about the "B" to the ICM not having voltage during cranking, that seems logical on its face. I don't know what that voltage is exactly supposed to be, but I assume it's an on-off signal at some voltage. I think checking the injector signals with a noid light was suggested, and that would be a useful next step. That would tell you if the PCM is sending a signal at all. If not, you may be back to the opti because it may be caused by the optical sensor not properly reading the wheel in there when the sensor gets hot. I hadn't thought about that when I wrote my last post.

KenMathisHD

wait, so then if the injectors aren't getting the signal then it may be the opti not sending the pulses, and if they are then it might be the pcm or something crimped/shorted along the wire running from the pcm to the icm? or will the opti not be ruled out even if the injectors are receiving a signal?
Sorry about all the questions, even with going through and reading posts and looking at the diagram, this system is kind of hard to understand :/

MatthewMiller

I'm no expert, and I agree this stuff isn't always the easiest to figure out even when you have the info. If the injectors are getting their signals, then I don't know all the possibilities it could mean. But if the injectors are getting their signal but neither the opti or coil are outputting any spark, then I don't see how it could be the opti since the coil is "upstream" of it in the spark sequence.

Renfield

You also happen to be at the outer reaches of the useful lifespan of the original Optispark.

When you DO get the noid light on the injector, you will have your answer. My wallet is on the Optispark. And you have my deepest condolences.

On the plus side, it does provide an opportunity to do a whole bunch of "while you're in there" stuff.

auggy

Try a new coil.

KenMathisHD

Today I finally had some time to mosey on down to the local autozone and have them scratch their heads when I mentioned noid lights

They finally found the set and I took off to the vette, where I hooked them up, and stupidly scalded myself on the bright silver metal parts of the ignition
Cranked her and... noid lights flashed!
But she wouldn't start....
I haven't cranked her since my last post, but whatever is unhappy is probably pretty dead now

So I checked measurements at the opti harness again to make sure it was all good
5volts dc @ A
5volts dc @ B
12volts dc @ C
0.26-0.3ohms @ D (original resistance is 0.2 so im not sure if this is significant enough)
All the above are with the negative cable connected to ground terminal D

On an odd note, when I connected positive cable to ground @ D and negative cable to A, the car acted like I had just reconnected the battery, but the battery was still connected(headlight motors whirred, there was also a click sound when connected). Again, I don't know if that's significant

So I went back to the ICM connector and tested values again
A got 10v
D got 10v
C is a ground, didn't touch that
Checked B while cranking, still no volts

At this point I got my buddy on the phone and talked with him about it and we wondered if maybe B was a shorted or open wire (there's a lot of that going around with this particular car), so I looked up a pinout diagram of the PCM, found the ignition control wire and we tested it
No resistance said my multimeter

So what I'm seeing so far, and please do correct me if I am wrong, is that the ICM isn't getting a signal from the PCM to tell the coil to spark, but the injectors are firing which means the PCM is getting a low res signal from the opti, which means the opti is working, but something in between is killing signal to the ICM, and whatever that is has completely failed now since the car won't start up anymore

This is as best as I can get to finding that the opti is sending a signal without hooking a Tech 1 up to the car (which I sadly dont have...)
Looking at the checklist, it sounds like the PCM is bad, but maybe there is something I am missing?

KenMathisHD

There isn't anything getting to the coil :/

Tom400CFI

YES. The distributor is the "starting point" or reference signal for all things that need a reference signal in the car;
*Ignition
*fuel pump
*tach
*injector pulses

If you have NONE of those things, it's the distributor or the wiring from the distributor to the ECM. If you have some of those functions...it's not the distributor.

You don't need a "Tech 1"...you can use any decent OBD I compatible scan tool. The fastest way to diagnose the opti is to see if you have a tach signal while cranking. If you don't...opti is probably not functioning.

HERE IS A THREAD where I PROVED that my opti was dying when hot. You could perform the same test.

KenMathisHD

A tach signal while cranking? Like the tach moves/bounces/whatever when cranking?

On a side note, if there's no signal to the ICM to get to the coil, would there be a tach signal?

The car still isn't turning over, so at this point the opti heat test is out. Is there any merit to the idea of the low res signal being too weak to get sent to the ICM but strong enough to get sent to the injectors? Cause I'm getting a signal for the injectors to fire and I have fuel pump activity still but no signal to the ICM

Btw, I actually have gone to your post a few times, I'm glad you ended up getting your vette up and running again! And I like the way you started the post in order to make people aware of the fact that the opti is blamed way too much

Sorry for the 20 questions, I've only had the car for maybe 3 months now and I'm still learning how it's systems work, it's a lot tougher than learning my truck lol

Please do let me know if any of the stuff I say is wrong, I'm eager to learn all I can with this car

MatthewMiller

I can answer this. Yes, because the tach signal comes from the PCM to the tach. It does not go through the ICM on its way. The ICM signal is a completely different output from the ECM. Here is a diagram of the ignition system from the actual service manual:



On the PCM block at the top right, you'll see that terminal #5 is the ICM signal output (same as in the Autozone diagram you pasted) and #13 is the engine speed signal output that will run the tach. So if the PCM is getting a good signal from the opti (terminals 2, 20, 14, and 3), it is possible that is running the tach, fuel injectors, etc., correctly; but the ICM may not be getting a signal. If that's the case, then the fault is either a break in the wire to the ICM (wire #423), or a fault in the PCM itself.

Just to be sure, when you say you are getting a signal to the injectors, you are seeing an alternating sequence between current and no current? You aren't just seeing current constantly on even with the engine cranking, right? And you're not watching them with a test light grounded somewhere else, are you?

Just since I was copying from the manual, here is the injector circuit:



So the injectors are fed constant voltage from the "Power Distribution Cell 10." That voltage is not switched on/off at that source - it is always available (unless fuse 22 or 23 blows!). After the injectors, the circuits run to the PCM, which controls them by grounding the circuit (injectors get current and will open to spray fuel) and disconnecting them from ground (injectors stop seeing current and therefore they close) every other crank rotation.

So your test light has to be in series with the injector circuit. If you connect one lead to the circuit and one lead to ground, you should always see a constant light. I guess you'd also being spraying a bunch of fuel into that cylinder, too...so make sure you aren't doing that!

PS: Are you getting any DTCs (trouble codes)?

PPS: Maybe Tom can chime in here if I'm wrong about the following. I'm not sure it is accurate to test for voltage at terminal B of the ICM. I believe that should be a 5v signal that comes on briefly four times each crank rotation to tell the ICM to fire the coil. If you just put a volt meter on it with the engine not cranking, you probably will not see voltage. I think you'd need a test light of some sort in series with this circuit (from the PCM to the ICM) and look for a flashing test light when the engine cranks. If you see the light flashing, you'd know the ICM is actually getting a signal from the PCM. Addendum: Read post #6 in this thread.

Tom400CFI

Matt covered a LOT there...and it's all fantastic information. I'll just add a couple things:
Yes that or using a scan tool would be better. Tach MAY not move, while cranking depending on how fast it cranks.




Not likely. As Matthew correctly stated, the signal from the distributor goes to the ECM, and that is the only place it goes. The ECM then uses that information to manipulate all the other actuators. I'd say that if you have fuel pump running while cranking, injectors firing...then the distributor is functioning -at least somewhat. Enough that the engine should fire at least.



Thanks!

KenMathisHD

Ahhh, ok. In the diagram I had it looked like the tach had something that connected through the coil somehow, it was quite confusing


Correct. I tested the injectors with a noid light and got pulses when she was cranking. I can't tell you how happy that made me to know that it might not be the opti

The only DTC's I am currently getting are related to my ABS wheel speed sensor, though that's because one of the sensors is broke still

I tested the ICM connector terminal B on AC while cranking looking to see if I got around 1-4v, but nada. It looks like B is the terminal that will get it's signal from the ECM while cranking/running but I'm getting nothing there so I may be wrong on that

KenMathisHD

Love posts like this! Somebody someday needs to make a website and consolidate posts that have steps like this!

I'm getting continuity on the white wire to the ICM but no volts going through it, so looks like either there's a bad connection in the PCM itself or the PCM is bad.

I know I've seen in some other threads that since our cars are kinda old that the solder joints may be subject to cracking, I know it happened to my wiper module on my 92 truck. I wonder if there's a way to crack the sucker open and resolder those joints and see if that will work? If I can do that I might just buy a new opti and pump anyways since she's @ 137k miles and idk if the PO ever mucked around with the opti
Then I've got a spare

And thanks a ton all of y'all! The info here is fantastic! I think I'm gonna screenshot these and keep them to use for the next problems