Overfueling issue: injector #6 not closing - FIXED!!

[Jeep_junkie]

Hi everyone, this is my first post here but I've used the forums a lot in the past for help with my JDM Aristo 2JZGTE install in a Jeep CJ7. It recently started running like complete crap and I'm a bit lost about how to fix it. The text below copied from my post on the 2JZ group on Facebook explains the issue in detail.

"I need help guys. My 2JZGTE is running extremely rough and rich. It hardly idles without misfiring and belching black smoke. Didn't just happen all of a sudden but rather built up to this point. Originally, it used to have a hard time starting when I first installed the motor. It would gradually catch and come up to idle speed and sometimes while cranking it would face occasional resistance like it was getting partially hydro locked. I figured dirty sticky injectors so I poured a bottle of injector cleaner into the tank and it seemed like the problem disappeared. However about a week ago it started the hard starting thing again with a steady miss that would go away within 10-15 seconds. Started it after a few days today and it ran like total shit, eventually just refusing to run at all. It is dead obvious that the injectors are just dumping fuel in the motor because it will start and run for a good 20-30 seconds with the fuel pump disconnected. No vacuum leaks, FPR is good, return line isn't kinked, took the plugs out and cleaned them just in case. They were dripping with gas. I can hear the injectors clicking when the motor is running but I'm still inclined to think that a couple of them might be stuck wide open. What else should I be considering??"

Since then, I have taken the fuel rail off the motor and tested it hooked up to the wiring harness and fuel supply. Injector #6 goes wide open as soon as the ignition key is turned on, even before the motor cranks. The other injectors pulse normally when the motor is cranked. However, crank position seems to have no effect on injector #6. It stays wide open for as long as the key is on. Turn the key off and it closes. I confirmed this right at the #60 terminal of the ECU also, so the wiring is not at fault. It's either the ECU itself, or something commanding the ECU to keep #6 open.

What should I be looking at now. The ECU's innards LOOK fine. I'm guess with a bad CPS none of the injectors would have worked right and neither would the ignition. What about camshaft position sensors? Any way to check them with a multimeter?

[jondee86]

Fuel injector control is done by the ECM providing a low voltage pulse to turn the power transistor ON for a few milliseconds. The power transistor then switches 12V to ground for a few milliseconds and the injector solenoid opens against the spring pressure. When the power transistor turns OFF, the spring closes the injector.

So either your power transistor has failed and shorted to ground, or something else is shorting the injector to ground any time you turn the ignition ON. Assuming your engine has sequential injection, there will be one transistor for each injector.

Cheers... jondee86

[2dogs]

Hi , Does the engine check lamp operate properly ?
If it works check the dianostic codes

Have you got a good multimeter and do you know how to use it ?


It is possible you have short to ground on the #6 injector ecu wire .

[Jeep_junkie]

So either your power transistor has failed and shorted to ground, or something else is shorting the injector to ground any time you turn the ignition ON. Assuming your engine has sequential injection, there will be one transistor for each injector.

Cheers... jondee86

Yes there is a transistor for each injector, NEC C4554 (NPN) to be precise. I located the one for injector #6 and did some resistance and continuity checks with a DMM. The readings for all 6 transistors are approximately the same. Nothing glaringly different for any one of them. So I guess that rules out the transistor itself. Therefore, it must be something upstream in the injector circuit. Does anybody have a detailed diagram for the injector circuit in the ECU? @wilbo666?

Hi , Does the engine check lamp operate properly ?
If it works check the dianostic codes

Have you got a good multimeter and do you know how to use it ?


It is possible you have short to ground on the #6 injector ecu wire .

Maybe you didn't read my original post. There were no DTCs, except for 34 which was set when I overboosted the turbos while adjusting the MBCs, so that's not related to the injector malfunction. I used a multimeter to narrow the issue down to the ECU itself. The wiring is good, injectors are good too.

[2dogs]

Hi Jeep junkie ,
Ok , seems that you know your way around with a DMM . I only saw this post .
From your tests I would agree that it is being caused by something else internally in the ECU . I have a modified JDM JZA80 2jzgte ECU here and some internal diagrams . I will dig it out over the weekend and see if i can help.
I dont suppose you can get hold of another ECU to run ?

[Jeep_junkie]

Hi Jeep junkie ,
Ok , seems that you know your way around with a DMM . I only saw this post .
From your tests I would agree that it is being caused by something else internally in the ECU . I have a modified JDM JZA80 2jzgte ECU here and some internal diagrams . I will dig it out over the weekend and see if i can help.
I dont suppose you can get hold of another ECU to run ?

I'm trying to score another ECU to compare this one with. Appreciate your help help mate

[Jeep_junkie]

Just a quick update. I decided to plug the ECU back into the harness, turn the key on and unplug everything on the engine one by one to see if the ECU terminal #60 (pin 15 in the 80 pin plug) would unground. Unplugging the Exhaust Gas Control Valve VSV ungrounds injector #6. Plugged in a spare VSV but no change so I'm now going to trace the wires in the harness. Fun and games LOL :/

[Jeep_junkie]

Well well it seems that the reason #60 is getting grounded is because pin 39, VSV2, is OUTPUTTING 1.9V, whereas the ECU actually internally grounds this pin to complete the EGCV circuit. This is probably the circuit I need to troubleshoot. For the sake of comparison, pins 38 & 40, VSV3 and VSV1 respectively, are not putting out any current at all. Now I need to figure out where on this circuit I might find a bad diode or transistor.

[wilbo666]

The only schematic that I have is one I traced myself for a 2JZGE. The business end of drivers circuits themselves are pretty simple (assuming GE is similar to GTE, which I would expect)...

While it is possible that there is an issue inside the engine ECU, I would think that the most likely cause would be a wiring problem, such as both the #60 and VSV2 wires having a fault (rubbed through?, etc) and both the #60 and the VSV2 wire shorted to ground...or another option is a short between the #60 injector wiring and the VSV2 wiring and the engine ECU turning VSV2 ON (To be honest I'm not sure what the exact logic for VSV2 is...).

A quick thing to do is to unpin the #60 pin and the VSV2 pin from the 80pin plug at the engine ECU and start measuring from there. That will tell you if the issue is in the engine loom, or the engine ECU... i.e. hook up a test LED to the now empty #60 pin directly at the engine ECU and see if it is ON all the time (if it is then that would point to an issue within the engine ECU...)?...also see if the injector stays ON with the #60 pin disconnected at the engine ECU?

If the issue is internal (I'd be a little bit surprised to be honest, but that happens from time to time! ) then you can start measuring the voltages at the base and collector of the #60 drive transistor inside the engine ECU during fault... is your theory that there is a fault within the engine ECU that is causing the #60 drive transistor to be turned on? or is your theory that there is a short within the engine ECU that is connecting the #60 terminal (before the #60 transistor collector) to the VSV2 terminal which is being grounded by the engine ECU? or something else?

Anyway if the issue is within the engine ECU then I think you'll have to slog through tracing the circuit with the multimeter would be my suggestion. I don't think there will be an easy way out (other than a replacement engine ECU). On the upside you will be able to use one of the other 5 working injector drivers as a "teachers edition"

Let us know how you get on.

Cheers
Wilbo

[Jeep_junkie]

Thanks for the details reply @wilbo666. VSV2 is specific to the GTE ECU. It opens the Exhaust Gas Control Valve which allows the second turbo to boost. The ECU does this by grounding the earth wire for the VSV inside the ECU. The positive side of this circuit is fed by the B and +B wires. All other VSVs are operated in a similar manner.

Now, I have ruled out any shorts or breaks in the wiring. I confirmed it is a problem in the ECU because whereas VSV2 is supposed to show 0v (it's an earth wire that isn't grounded until the ECU commands it), it is actually showing 1.9v when power is connected to the ECU. When the ECU and everything else is plugged into the harness, the particular VSV in question is receiving 12v power from one side but while it should have 0v on the other side, the ECU is trying to feed it 1.9v. THAT is causing the ECU to ground the #60 pin for the injector. Therefore, the driver circuit for injector #6 itself isn't faulty. It's just getting activated due to current leakage on VSV2. I hope that explains the situation a bit better.

I have discovered a busted capacitor in the ECU. Will try to post a picture of it in a bit. I hope I'm on the right track with this.

[wilbo666]

Hi Heep_junkie,

I think you must have misinterpreted some of my last post.

The only mention I made to GE was of the injector driver circuit, the rest of my comment was specific to GTE.

I'm aware of the EGCV, it's VSV and it's operation, when I was referring to not knowing the exact logic I was meaning when the engine ECU decides to open and close it exactly...i.e. when the engine is not running does the engine ECU energize this VSV for some strange reason? (All that said I was actually thinking about the VSV for the exhaust bypass valve, not that it really matters! )

Now, I have ruled out any shorts or breaks in the wiring. I confirmed it is a problem in the ECU because whereas VSV2 is supposed to show 0v (it's an earth wire that isn't grounded until the ECU commands it), it is actually showing 1.9v when power is connected to the ECU. When the ECU and everything else is plugged into the harness, the particular VSV in question is receiving 12v power from one side but while it should have 0v on the other side, the ECU is trying to feed it 1.9v.

I'm not sure that I'm following 100% correctly Can you tell me how exactly have you ruled out the engine wiring loom being involved?

THAT is causing the ECU to ground the #60 pin for the injector. Therefore, the driver circuit for injector #6 itself isn't faulty. It's just getting activated due to current leakage on VSV2. I hope that explains the situation a bit better.

I'm also not sure that I follow your logic here? Can you explain in more detail / in a different way?
How is a voltage on the VSV2 pin is causing the engine ECU to ground the #60 pin?

My understanding is that the #60 and VSV2 circuits are separate, relatively unrelated circuits! (other than both being inside the engine ECU...)

Are you able to answer the following questions for me?
1) How have you ruled out an issue in the wiring loom?

2) How is a voltage on the VSV2 pin causing an issue on the #60 pin of the engine ECU?

3) If you un-pin the #60 pin at the 80 pin plug of the engine ECU, does the injector remain stuck ON? (leaving all the other wiring as it should be, i.e. everything else connected, etc)


Cheers
Wilbo

[Jeep_junkie]

Hi Heep_junkie,

I'm aware of the EGCV, it's VSV and it's operation, when I was referring to not knowing the exact logic I was meaning when the engine ECU decides to open and close it exactly...i.e. when the engine is not running does the engine ECU energize this VSV for some strange reason? (All that said I was actually thinking about the VSV for the exhaust bypass valve, not that it really matters! )

Oh I see, my bad. I'm not sure of the exact logic used by the ECU either, just that the second turbo comes online between 3800-4000 rpm. So engine RPM is definitely part of the equation, as must TPS position and the MAP signal. I'm quite sure it's a lot more complicated than that too LOL.

I'm not sure that I'm following 100% correctly Can you tell me how exactly have you ruled out the engine wiring loom being involved?

I disconnected the wiring harness from everything on the motor (sensors, injectors, VSVs, etc) as well as the ECU. Then I used a DMM to look for any wires shorted to ground or to each other, in particular #60 going to the #6 injector and VSV2 going to the EGCV VSV. My tests confirmed that no wires are shorted to each other or to ground.

I'm also not sure that I follow your logic here? Can you explain in more detail / in a different way?
How is a voltage on the VSV2 pin is causing the engine ECU to ground the #60 pin?

My understanding is that the #60 and VSV2 circuits are separate, relatively unrelated circuits! (other than both being inside the engine ECU...)

Quite frankly, I wasn't expecting another circuit to be involved either. When I checked the ECU with only IGSW, MREL, B, +B, BATT, E1, E01, and E11 terminals connected to the battery (basically the power circuit for the ECU, nothing else attached) #60 did NOT become grounded. That led me to believe there was something in the wiring harness that was maybe feeding back voltage in to the ECU causing it to malfunction. Hence, I connected #60 directly to my DMM which was set to beep for continuity, and attached the other probe of the DMM to ground. Then I turned on the power to the ECU and began disconnecting the wiring harness from everything on the engine. The beeping stopped when I reached the EGCV VSV connector, so I knew there was something fishy on this circuit. I then disconnected the ECU also and checked the #60 and VSV2 wires in the wiring harness for shorts between the two or to ground. That ruled out the wiring harness. I did the same check again for the positive side of #60 wiring and VSV2 wiring - again, nothing wrong there.

Next step, I reconnected the ECU back to power only (IGSW, MREL, B, +B, BATT, E1, E01, and E11 terminals) and used the DMM to monitor the VSV2 pin. That confirmed the presence of 1.9v at VSV2. For the sake of comparison, I tested its neighbors, VSV1 and VSV3. Both showed no voltage under the same conditions, as I expected with only power being supplied to the ECU.

Are you able to answer the following questions for me?
1) How have you ruled out an issue in the wiring loom?

2) How is a voltage on the VSV2 pin causing an issue on the #60 pin of the engine ECU?

3) If you un-pin the #60 pin at the 80 pin plug of the engine ECU, does the injector remain stuck ON? (leaving all the other wiring as it should be, i.e. everything else connected, etc)


Cheers
Wilbo

1) Yes, as described above.

2) I can't really answer how as I do not know about the power flow between the circuits inside the ECU, so I am just as surprised as you in this regard.

3) No, the injector turns off when the #60 terminal is de-pinned. I verified this with the fuel rail removed from the engine but connected to the fuel system and injector harness.

I also found these two signs of capacitor leakage inside the ECU last night. Although I could not detect any continuity between the caps and the #60 and VSV2 terminals, there is continuity between the encircled blue resistor with the black smudge and VSV2.
IMG_20140803_231711.jpg
IMG_20140804_002942.jpg

[wilbo666]

3) No, the injector turns off when the #60 terminal is de-pinned. I verified this with the fuel rail removed from the engine but connected to the fuel system and injector harness.

OK, so if the injector turns off when the #60 terminal is de-pinned (and all the other engine ECU terminals are still connected), then that would point to the issue being within the engine ECU as you have said (I'm still quite surprised to be honest! )

The next thing I would be doing is checking where along the line the #60 terminal is being grounded, i.e. measure the voltage at the #60 transistor base... unsolder the #60 transistor and see if the collector is grounded... etc (It sounds like you've got a pretty good handle on fault finding, so prbly don't need my help here ).

I also found these two signs of capacitor leakage inside the ECU last night. Although I could not detect any continuity between the caps and the #60 and VSV2 terminals, there is continuity between the encircled blue resistor with the black smudge and VSV2.

Leaking caps within older Toyota ECUs is quite common unfortunately. Defiantly a good idea to replace. Ear cleaners dipped in acetone are good for cleaning up the capacitor residue / conformal coating / flux left over from when you solder the new caps in also FYI.

Cheers
Wilbo

[Jeep_junkie]

OK, so if the injector turns off when the #60 terminal is de-pinned (and all the other engine ECU terminals are still connected), then that would point to the issue being within the engine ECU as you have said (I'm still quite surprised to be honest! )

The next thing I would be doing is checking where along the line the #60 terminal is being grounded, i.e. measure the voltage at the #60 transistor base... unsolder the #60 transistor and see if the collector is grounded... etc (It sounds like you've got a pretty good handle on fault finding, so prbly don't need my help here ).



Leaking caps within older Toyota ECUs is quite common unfortunately. Defiantly a good idea to replace. Ear cleaners dipped in acetone are good for cleaning up the capacitor residue / conformal coating / flux left over from when you solder the new caps in also FYI.

Cheers
Wilbo

I did some basic tests on the #60 transistor as well as the transistors for the other injectors and they all checked out fine. I'm going to trace the circuit fit VSV2 and see if there is anything else besides the bad caps that needs to be replaced.

As of now I am driving the car again. It runs perfectly fine with the EGCV VSV unplugged from the ECU.

[Jeep_junkie]

Well I'm back with not so good news. I had a friend who's an electronics engineer to have a look at it and he spotted a couple more caps that were leaking. However, he was unable to find replacement parts so he just soldered them back in and returned the ECU to me. He said the leaked electrolyte may have been causing problems so just cleaning the mess could even fix the problem. I decided to try out the ECU and the check engine light wouldn't come on because the ECU is sending out 12v on the W pin which it is supposed to ground when the ECU gets power. Similarly, other pins which should be grounded internally by the ECU, such as the VSVs, are also outputting 12v. Please help!