DBJoe996

Throughout all this, and I know that just throwing in parts is never a really good idea, but to start to eliminate some things have your mechanic install NEW coils and NEW spark plugs now. You have to start getting a baseline for things. Get pictures of the condition of each cylinder spark plug and post. You can tell a lot about a cylinder by how the spark plug looks.

wildbilly32

I find the above strange...anyone else? Unless this is a weak point for these ECUs what's the odds that two would be failed in the same manner? In addition you didn't believe Specialized on ECU A? I'm with DB above plugs and coils inspect as they come off. Verify injectors are fresh and of proper type(I might have missed if this already done). PT3 above with his "it's usually the simple things" post is excellent information, as well. Good Luck.

Porschetech3

You also have to remember that in the scientific world, electron flow is actually opposite of what is commonly known as conventional current flow, in that electrons flow from negative battery post toward the positive post.

So for the injector to work, the electrons flow from the negative battery post, into the metal chassis, then through GP9 and GP8 and into the DME.Then out of the DME to inj.#1 or inj. #5 ect. and to the 12v splice and back to the battery positive.

Since injectors require the highest amount of electrons to work ( of the DME curcuits), if there is not enough flow of electrons through GP9 or GP8 at high load (high injection time) the injectors will be erratic. There could be slight differences in the internal resistance paths in the 3 different DME's that if there were enough electron flow these slight differences would not effect the injector function.

It's also common knowledge that bad ground point connections get worse with heat or load/usage time, and can cause damage to DME curcuits.

Of course these are just theories, and could be enlightening if it proves to be the problem..

wildbilly32

Yeah, in a past life, before entering management, I was a railroad electrician so I'm aware of electricity. I agree wholehearted with the simple things theory you are promoting i.e. your connection checking/cleaning suggestions. I saw way too many "hot shot" locomotive troubleshooters run around in circles convinced the solid state devices they had changed numerous times must all have been bad and when left for me to fix I found a broken wire on a terminal board that was hiding behind a cover in about 20 minutes after a week lost to the "experts". Supervisor thought I was lying to him when I told him it was fixed and what should I do now. Simple things...

Imo000

Sounds to me the easiest thing to do would be to connect one end of the jumper cable to the negative battery post and the other end to the engine block and see what happens.

Porschetech3

Doing that would not help the current flow through the DME and injectors.

Imo000

Ok then find the pin for the DME ground and extend it to the battery negative. That should do it. What I'm trying to get to is to make sure the supplied ground is good.

808Bill

Seen many electrical issues solved by a better ground.

Porschetech3

Gotcha, that would work, the DME has several ground wires..

Paul Waterloo

That is very odd indeed. Based purely on what you have written, it sure seems like an electrical problem, and if I had to guess, something with a bad electrical connection, which could be a ground, a high resistance connection outside of the ground, or maybe even a rodent that ate part of your wiring harness. Sorry I can't be much more help, but I sure don't think you have three bad ECU's. It sounds like you have three good ones.

FlyingShawn

Thanks for the continued input, everyone! I emailed my indy late last night with some of your thoughts and a handful of other things I learned during many, many hours of searching and reading on here (including some excellent info @Porschetech3 gave to users in some other threads, such as the cat protection logic in the DME that would explain why the #1 injector is getting shut off completely!), but I haven't heard back from him about what he was able to try today or the results.

Based on the growing consensus here I'm getting more and more confident that we need to focus our efforts on looking for issues elsewhere in the electrical system beyond the fuel injector circuits we've spent so much time on.

Besides all this great electrical discussion, the two components that are still on my radar are the coils and the MAF. One interesting thing that turned up in my research was the idea to do a test drive with the MAF unplugged: theory being that if the MAF is the culprit then the misfire should go away without any MAF input data (engine would be less powerful during the test, of course, but if the misfire persists we'd be able to scratch out the "bad MAF" theory).

I'll keep you updated as things develop!

FlyingShawn

Just heard back from my Indy. Here's what he said (slightly edited for clarity, such as using our ECU A-C nomenclature):

Besides scoping the injector signal waveform, any ideas? I suggested again that we try the MAF-unplugged test, but that's just a shot in the dark. Also suggested @Ahsai 's idea of disconnecting/cleaning the X59 connectors.

Like @Porschetech3 I'm concerned that we might be fixating on the DME and not thinking outside of the box enough, but that test of running new grounds to the DME seems like it would have eliminated a lot of possibilities.

FlyingShawn

Stopped by the mechanic yesterday. Apparently he never got my reply a few days ago (must have been a Gmail glitch because it's also not anywhere to be found in my sent or draft items), so we spent some time working on a a gameplan. We're going to:

-Replace a cam sensor that's leaking oil (we've tried cleaning and reinstalling it already, but the leak is back). Shouldn't make any difference for our misfires, but it still needs doing and what's weird about it is that it's somehow wicking oil into the X59 connector via the harness wiring. Oil is an insulator, so it shouldn't(?) make any difference for our misfire issues, but it's still weird. We'll re-clean the X59 if more oil has found its way in there since when we first cleaned the cam sensor a few weeks ago.

-Replace the other four coils. We've already done our problem cylinders #1 and #5, and we're not expecting this to make any difference for our misfires, but it makes sense to do the rest just to have them done. I'll confess I'm getting frustrated enough that I'm more willing to throw money at the thing just in the hopes that something would happen, especially as we're getting into top-down weather and my mechanic has put more miles on the thing than I have!).

-Swap the #1 and #2 injectors to see if the problem moves with ECU C installed

-Test drive it with the MAF unplugged

-Record the waveforms of the fuel injector signal pulses from the ECU

-And, lastly, if none of the above makes any difference, our last idea apart from a 4th DME would be to start unplugging/examining/cleaning every electrical connector we can get our hands on even if it doesn't seem remotely related to the problems we've been having. I've already told you that he's tried bypassing the DME grounds, but I didn't know until yesterday that he's also cleaned and bypassed the main engine ground as well.

Any other out-of-the-box ideas we should consider? Is there any conceivable way that the oil in the harness from the cam sensor could be causing all of this despite oil being an insulator?

Flat6 Innovations

First off, when I start chasing any issues, I stop and verify mechanicals immediately.
I'd say a running compression test is in order for this one, before you chase your tail any further.

I have seen injector codes and misfires from cylinders that had no injector, or coil faults, but had mechanical issues on that same cylinder. A running compression test is the only way to go here, static tests won't show the issues.

FlyingShawn

The reason we haven't considered a compression test is that each of the "misbehaving" cylinders function flawlessly with one DME or the other. ECU A and cylinder #1 runs great (but cylinder #5 has problems after a while), ECU C and cylinder #5 runs great (but #1 starts acting up shortly after engine start). Wouldn't a mechanical issue with a given cylinder misbehave regardless of which computer is installed?