Resolved 2GNT ECU failing Intermittently?

[Nova95]

I've been chasing a weird problem with my 96 Eclipse 420a NT for a few months now. Randomly in the middle of driving she'll cut out and will not start back up. Eventually it will start again on its own. Usually after killing my battery first, and having someone jump me for 5-10.

Lately it's gotten worse and she just will not restart. I've replaced the cam/crank sensors, fuel pump, filter, MAP, Ignition switch, coil pack, plugs, wires. With each part I thought I had fixed the problem because it would run perfect for the next two days or so then I'd cut out in the middle of traffic again.

The fuel and ASD relays are good, but the connectors were pretty badly corroded so I replaced both of them. It ran good for a day then same thing. Unplugging the battery for 5-10 minutes allows me to restart it, but then she dies even if I try to gas it as if the fuel just got cut.

Yesterday and today she died in some pretty crappy spots and the only I was able to limp her home was by jumping the ASD relay to the negative on my battery. She tried to cut out but both times I could actually rev her up to 5-6k to stop her from cutting out (finally getting fuel). After about 5 minutes of driving in these RPMs, the car acts perfectly normal again like nothing ever happened. I can unplug the jumper wire and use the car as normal for the rest of the day till it decides to die randomly again, usually the next day.

Oh and the car is not throwing a single code for me to even try to work with. What is going on? Could my ECU be failing just sometimes?

 

[danl]

What is the ASD relay?

First thing is to determine if you are losing spark or fuel when this happens.

 

[luv2rallye]

Ya the ECU may be going out (typically bad leaking capacitors - rotten egg smell when you open it and crap leaking around some capacitors on the circuit board). But first check that the battery is tied down with insulator over the positive terminal so it doesn't short against the hood which it can then do. Also check that all the air intake ducts clamps are tight so no leaks.

 

[Nova95]

What is the ASD relay?

First thing is to determine if you are losing spark or fuel when this happens.

It's the relay to the right of the fuel pump relay. It tells the ECU to kill the engine due to failing parts is what I've gathered to be the jist of it but all of the parts associated with it I've replaced at the point. It's definitely the fuel getting cut, spark is still there when this happens on all 4 cylinders.

What is the ASD relay?

First thing is to determine if you are losing spark or fuel when this happens.

It's the relay to the right of the fuel pump relay. It's definitely the fuel getting cut, spark is still there when this happens on all 4 cylinders

Ya the ECU may be going out (typically bad leaking capacitors - rotten egg smell when you open it and crap leaking around some capacitors on the circuit board). But first check that the battery is tied down with insulator over the positive terminal so it doesn't short against the hood which it can then do. Also check that all the air intake ducts clamps are tight so no leaks.

I've triple checked all of the intake definitely nice and tight. To know if anything is leaking with the ECU would I remove it and open it up? I've only pulled the two harnesses coming out of it and everything looked fine enough.

It's the relay to the right of the fuel pump relay. It's definitely the fuel getting cut, spark is still there when this happens on all 4 cylinders

I've triple checked all of the intake definitely nice and tight. To know if anything is leaking with the ECU would I remove it and open it up? I've only pulled the two harnesses coming out of it and everything looked fine enough.

If I come to find that the ECU is the culprit, will grounding out pin 4 the way I have at least keep the car going to and from work until the new ECU comes on or will I still most likely die in traffic?

 

[BLACK'98DSM]

Do you have a valve cover leak?
I ask because the main fuel injector harness connector sits just below the valve cover, on the naturally downhill side, and if you have a leak it will fill the connector with oil and cause shorts or loss of connection. Though it's a sealed connector, all of the old seals are bad now. I've seen this before. Check here first.
This would even explain the intermittent aspect of your issue. Over time the oil will drain from the connector. The vibration of engine rotation (turning over with starter) will catalyze the oil drain. Once the pins are cleared enough to make connection, it will start and run again until the connector has refilled with oil. Repeat process.

Now, visually identifying a bad ECU component or circuit is usually not easy. And sometimes is not possible. The entire circuit board assembly is coated with a thick layer of gel, to reduce vibrational damage and prevent corrosion. It can be tough to get a detailed look at the board components since the gel has a hazy brown tint. Note, there are intricate ways to test each function of the ECU. But this is not enticing with an intermittent issue. If no issues are found with the injector harness or valve cover, it may be easier to just swap with a used ECU than to try diagnosing yours. I do think it's worth opening your ECU to inspect, but a small issue like a leaky capacitor will be hard or impossible to pinpoint. Unless something has really burned, it likely won't show through the gel.

 

[luv2rallye]

If the ECU is the culprit, nothing you do but fixing the ECU will fix it.

The ASD relay does not "kill the engine due to failing parts". Instead it is the main relay that supplies power to the ECU (called a PCM on a 420a), coils, O2 sensors, and injectors. It would be dangerous to leave the ASD forced on (leaving pin 4 shorted to ground assuming the relay works) when not driving the car. Only do that in an emergency to get home and remove jumper when done - but that assumes the relay is working. The ASD relay itself may also be the problem. To eliminate the relay, unplug it and short its pin 8 (larger red-black) to its pin 2 (larger black-red) with a large wire, which is what the relay does - but only do this as a test - UNDO THIS when done running the engine!

If the engine now runs with no problems, the ASD relay itself is most likely the problem.

 

[Nova95]

Do you have a valve cover leak?
I ask because the main fuel injector harness connector sits just below the valve cover, on the naturally downhill side, and if you have a leak it will fill the connector with oil and cause shorts or loss of connection. Though it's a sealed connector, all of the old seals are bad now. I've seen this before. Check here first.
This would even explain the intermittent aspect of your issue. Over time the oil will drain from the connector. The vibration of engine rotation (turning over with starter) will catalyze the oil drain. Once the pins are cleared enough to make connection, it will start and run again until the connector has refilled with oil. Repeat process.

Now, visually identifying a bad ECU component or circuit is usually not easy. And sometimes is not possible. The entire circuit board assembly is coated with a thick layer of gel, to reduce vibrational damage and prevent corrosion. It can be tough to get a detailed look at the board components since the gel has a hazy brown tint. Note, there are intricate ways to test each function of the ECU. But this is not enticing with an intermittent issue. If no issues are found with the injector harness or valve cover, it may be easier to just swap with a used ECU than to try diagnosing yours. I do think it's worth opening your ECU to inspect, but a small issue like a leaky capacitor will be hard or impossible to pinpoint. Unless something has really burned, it likely won't show through the gel.

Actually the back side of the valve cover area is leaking oil, I hadn't thought of that even being a problem. I'll go see how the harness looks and clean it out with electrical cleaner. I just now put the ECU back together, I didn't see anything through the gel that looked burned up, but I also am not confident that there couldn't be as I could barely see through the gel, I did clean up the dust a bit though. I'll check the harness and post back with what I find.

If the ECU is the culprit, nothing you do but fixing the ECU will fix it.

The ASD relay does not "kill the engine due to failing parts". Instead it is the main relay that supplies power to the ECU (called a PCM), coils, O2 sensors, and injectors. It would be dangerous to leave the ASD forced on (leaving pin 4 shorted to ground assuming the relay works) when not driving the car. Only do that in an emergency to get home and remove jumper when done - but that assumes the relay is working. The ASD relay itself may also be the problem. To eliminate the relay, unplug it and short its pin 8 (larger red-black) to its pin 2 (larger black-red) with a large wire, which is what the relay does - but only do this as a test - UNDO THIS when done running the engine! If the engine now runs with no problems, the ASD relay itself is most likely the problem.

The first thing I tried when it died was pin 8 to pin 4, it still wouldn't start. When I pulled the jumper and put it from ground to pin 4 it starts and gets me home. I've also tried swapping the relays before I even did the jumper to make sure it wasn't a bad relay before I even dared drive home on the jumper.

 

[luv2rallye]

The first thing I tried when it died was pin 8 to pin 4, it still wouldn't start. When I pulled the jumper and put it from ground to pin 4 it starts and gets me home. I've also tried swapping the relays before I even did the jumper to make sure it wasn't a bad relay before I even dared drive home on the jumper.

If you shorted the real ASD pin 8 to the real ASD pin 4 (assuming you know which is the real pin 8 and 4 since the connector is complicated), that's a complete short going to the ECU's pin 67. You may have just blown your ECU's output driver transistor on pin 67 if the ECU tried to turn on (because the ECU would be missing the necessary resistance of the ASD's coil).

 

[chrysler kid]

back in the old days the 420a motors in the neon and pt cruiser etc use to suffer from crank sensor failure due to a bad connection at the pigtail. typically and oil leak from the valve cover gasket would find its way into the crank sensor pigtail. I believe at one point there was a technical service bulletin that stated the crank sensor should be replaced with the pigtail for the crank sensor

 

[Nova95]

If you shorted the real ASD pin 8 to the real ASD pin 4 (assuming you know which is the real pin 8 and 4 since the connector is complicated), that's a complete short going to the ECU's pin 67. You may have just blown your ECU's output driver transistor on pin 67 if the ECU tried to turn on (because the ECU would be missing the necessary resistance of the ASD's coil).

My apologies, I meant I put the jumper from 8 to 2 with no results. Only after

If you shorted the real ASD pin 8 to the real ASD pin 4 (assuming you know which is the real pin 8 and 4 since the connector is complicated), that's a complete short going to the ECU's pin 67. You may have just blown your ECU's output driver transistor on pin 67 if the ECU tried to turn on (because the ECU would be missing the necessary resistance of the ASD's coil).

Sorry typo. I shorted pin 8 to pin 2. (30 to 87). It didn't start. I shorted pin 4 to ground and the relay clicks and it starts. I've also just noticed with key in the on position the ASD relay does not click. Both fuel and ASD should click with key in the on position right? I've swapped the fuel and ASD relays to see if it's the relay and the ASD does not click under either. Swapped the fuel pump under both relays and it clicks on under both relays with no problem.

 

[Nova95]

Okay I double check all of the connectors around the bay that could of had oil in them, all looked fine. I cleaned them all up with electrical cleaner just to be sure and squeezed them nice and tight anyhow. Car started right up today, and drove to the grocery store, but I won't know if anything has really changed until I go back to work Wednesday and have to use the car off and on all day.

 

[luv2rallye]

I shorted pin 8 to pin 2. (30 to 87). It didn't start. I shorted pin 4 to ground and the relay clicks and it starts. I've also just noticed with key in the on position the ASD relay does not click. Both fuel and ASD should click with key in the on position right? I've swapped the fuel and ASD relays to see if it's the relay and the ASD does not click under either. Swapped the fuel pump under both relays and it clicks on under both relays with no problem.

Perhaps I'm wrong but all this keeps telling me the ECU is the problem. The ECU is the thing that should activate the ASD relay and it's not (you are forcing it). There's more than just activating the ASD relay to start the engine - the ECU itself must be in a starting mode (which I don't think it is) to control all kinds of things. Shorting pin 8 to 2 does what the ASD relay would have done but it doesn't put the ECU in start mode. Shorting pin 4 to ground is what the ECU would have done if it was in start mode, but you did it, which I suspect improperly forced the ECU into a half assed start mode so it started. But it's not the correct way to get the ECU into start mode (and may even be damaging to the ECU - eg. you forced a ground on pin 67 where normally it floats and the ECU controls it - not an outside force). And now the ECU may still not be completely in start mode for all functions (eg. some of the ECU internal circuitry that ends up bringing pin 67 low is still not in its correct state, and that circuitry may be driving other needed things).

Also note that ECU's aren't always all good or all bad. They are ruggedly built to withstand all kind of things and so can get intermittent. Even when they start failing, they are designed to operate as best as they can with what's failing. Another thing you can check if to make sure you have constant voltage on the battery backup pin 46. If not I've heard stories of it causing intermittent problems even though it's only supposed to be for ECU parameters backup.

 

[Nova95]

Perhaps I'm wrong but all this keeps telling me the ECU is the problem. The ECU is the thing that should activate the ASD relay and it's not (you are forcing it). There's more than just activating the ASD relay to start the engine - the ECU itself must be in a starting mode (which I don't think it is) to control all kinds of things. Shorting pin 8 to 2 does what the ASD relay would have done but it doesn't put the ECU in start mode. Shorting pin 4 to ground is what the ECU would have done if it was in start mode, but you did it, which I suspect improperly forced the ECU into a half assed start mode so it started. But it's not the correct way to get the ECU into start mode (and may even be damaging to the ECU - eg. you forced a ground on pin 67 where normally it floats). And now the ECU may still not be completely in start mode for all functions (eg. some of the ECU internal circuitry that ends up bringing pin 67 low is still not in its correct state, and that circuitry may be driving other needed things).

That makes sense to me. I'll work on getting an ECU replacement to see if it changes anything. But back to the ASD relay for a sec, If I turn the key to the on position, is the relay supposed to click for a second just like the Fuel relay does?

 

[luv2rallye]

That makes sense to me. I'll work on getting an ECU replacement to see if it changes anything. But back to the ASD relay for a sec, If I turn the key to the on position, is the relay supposed to click for a second just like the Fuel relay does?

Only if the ECU tells it to since it controls it. I would think the ECU tells it to stay on or stay off - not just click on for a second. But admittedly I'm more knowledgeable with the turbo engine then the NT (chrysler has their own ways). On the turbo the CEL goes on for a few seconds telling you the ECU is in start mode. Do you normally get that (before all these problems)?

 

[Nova95]

I checked all pins of the ASD relay with key in both on and off positions, these look fine correct?

KeyOff Pin 2: 12.85. KeyOn Pin 2: 12.65
Pin 4: 1.34. Pin 4: 2.68
Pin 6: 12.85. Pin 6: 12.65
Pin 8: .006. Pin 8: .008

Just tried to make a second trip and it died, jumpered and drove back home just fine. I guess it's getting parked till I can get another ECU. Thanks for the help guys.

 

[luv2rallye]

I checked all pins of the ASD relay with key in both on and off positions, these look fine correct?

KeyOff Pin 2: 12.85. KeyOn Pin 2: 12.65
Pin 4: 1.34. Pin 4: 2.68
Pin 6: 12.85. Pin 6: 12.65
Pin 8: .006. Pin 8: .008

The NT pages of my 1999 manual shows pins 6 and 8 are tied together so your pin 8 probably has a wiring problem.

Just tried to make a second trip and it died, jumpered and drove back home just fine. I guess it's getting parked till I can get another ECU. Thanks for the help guys.

Hold off on getting another ECU (since they are expensive) until you exhaust all other possibilities first. Also maybe getting a NT expert (eg. Doug99RS, Black'98DSM) to help with this thread may be prudent. Even Steve who is a turbo expert knows ECU operation better than anyone so may know the NT too.

 

[Nova95]

The NT pages of my 1999 manual shows pins 6 and 8 are tied together so your pin 8 probably has a wiring problem.

Crap I'm sorry I wrote down the pins incorrectly. This is my first time really looking at wiring diagrams and it's a lot on my brain . What you said is correct. My 6 and 8 are tied together. Where would I look in the engine bay for the ground for the ECU? Back of the firewall?

When the car starts, it idles perfect and will idle for as long as I have it on. But about 5 minutes into the actual driving it when this is happening now. It sputters/jerks once or twice briefly, and that's when I know to pull over because any Minute after that second sputter, I lose it.

 

[luv2rallye]

Ya that typically sounds like a lack of fuel problem. Have you checked fuel pressure (which should be maintained for days) or for a leaking fuel pressure regulator?

Where would I look in the engine bay for the ground for the ECU? Back of the firewall?

I'm not an NT guy but it's wherever ECU pins 10 and 50 go.

 

[Nova95]

I want to say it was 50psi at when I replaced the pump and filter. I'd have to rent a gauge again to double check it's still at that.

 

[luv2rallye]

Yes but does the fuel pressure stay at the rail for days (which it should) or does it bleed down in just minutes/hours (eg. due to a leaking fuel pressure regulator, or leaking fuel pump check valve, or leaking injectors, etc)? Also does the fuel pump maintain proper pressure while driving (or might that be your problem)?

 

[Nova95]

I'm guessing id need quite a better obd2 than what I have to know what my pressure is while I'm driving. Is the regulator next to the fuel filter? Probably cheaper for me to just replace and cross off since the rest of the fuel system has been replaced at this point.

I've been testing the car up and down my street with and without the ASD ground jumped and pretty much I'm at the car won't get passed leaving my complex with no jumper, stalls, and will just crank till I kill battery in the driveway. Now the second I throw the ground jumper in, the car starts instantly. If I start the car already with jumper it will drive about 10-15 minutes sometimes more before cutting out. At this point I then have to pull the negative on the battery, for a second then put it back and the car will drive again with no problems for another 10-15 minutes. This makes zero sense to me but most of what I know about cars is on the mechanical side of things. Electrical is a lot.

 

[Nova95]

How long is an ECU expected to last anyways? My car just hit the 200k club when all of this mess started.

 

[luv2rallye]

I'm guessing id need quite a better obd2 than what I have to know what my pressure is while I'm driving. Is the regulator next to the fuel filter? Probably cheaper for me to just replace and cross off since the rest of the fuel system has been replaced at this point.

On the turbo you have to attach a fuel pressure gauge to the fuel rail (or fuel line) to read it. I assume the NT is the same - I doubt you can read it via the obd2 since I doubt there's any fuel pressure sensor on the NT engine. But then I've never had a NT.

Yes on the NT my manual shows the FPR is attached to the fuel filter. I agree it's easier to just replace it as it's not expensive and probably old enough to need replacing anyway.

I've been testing the car up and down my street with and without the ASD ground jumped and pretty much I'm at the car won't get passed leaving my complex with no jumper, stalls, and will just crank till I kill battery in the driveway. Now the second I throw the ground jumper in, the car starts instantly. If I start the car already with jumper it will drive about 10-15 minutes sometimes more before cutting out. At this point I then have to pull the negative on the battery, for a second then put it back and the car will drive again with no problems for another 10-15 minutes. This makes zero sense to me but most of what I know about cars is on the mechanical side of things. Electrical is a lot.

When you throw the jumper on you're forcing the ASD relay to activate which powers the alt field coil, ignition coils, O2's, injectors, and tells ECU the ASD has activated (which it needs to know to start controlling engine things like spark timing and injector pulse width and mixture). Ya, pulling the battery negative to get it going again is strange, other than that does an ECU reset which apparently gets you going for a while with whatever the problem is. The problem may indeed be an intermittent ECU. I just wanted to make sure it wasn't low fuel pressure because your symptoms fit that too the way it dies out on the road (and I've seen sometimes after the engine dies and you wait a couple minutes {like perhaps the time to pull the battery negative}, enough low pressure fuel works it way into the throttle body to allow it to idle or start up again and the sequence repeats).

This certainly is a strange one.
Have you replaced the fuel filter? A clogged filter can sometimes do this.

Can you race the engine while not moving without it quitting/missing proving it's getting enough fuel?

It's also possible something is failing with temperature (even the ECU). Heating components with a hair dryer with engine running can sometimes bring out temp issues.

Know that if it is the ECU, ANYTHING weird is possible.

The ECU when it decides everything is ok (getting proper signals in it's power up sequence) is the thing that activates the fuel pump (driving ECU pin 74 low) and ASD relay (driving ECU pin 67 low). So if the ECU doesn't see everything is proper it won't activate these. If that is the case when you are forcing the ASD, you can see the incompatibility with that and you would need to figure out what is wrong in the ECU power up sequence (keeping in mind the ECU itself may even be the problem). Or just try a different ECU.

 

[Nova95]

On the turbo you have to attach a fuel pressure gauge to the fuel rail (or fuel line) to read it. I assume the NT is the same - I doubt you can read it via the obd2 since I doubt there's any fuel pressure sensor on the NT engine. But then I've never had a NT.

Yes on the NT my manual shows the FPR is attached to the fuel filter. I agree it's easier to just replace it as it's not expensive and probably old enough to need replacing anyway.

When you throw the jumper on you're forcing the ASD relay to activate which powers the alt field coil, ignition coils, O2's, injectors, and tells ECU the ASD has activated (which it needs to know to start controlling engine things like spark timing and injector pulse width and mixture). Ya, pulling the battery negative to get it going again is strange, other than that does an ECU reset which apparently gets you going for a while with whatever the problem is. The problem may indeed be an intermittent ECU. I just wanted to make sure it wasn't low fuel pressure because your symptoms fit that too the way it dies out on the road (and I've seen sometimes after the engine dies and you wait a couple minutes {like perhaps the time to pull the battery negative}, enough low pressure fuel works it way into the throttle body to allow it to idle or start up again and the sequence repeats).

This certainly is a strange one.
Have you replaced the fuel filter? A clogged filter can sometimes do this.

Can you race the engine while not moving without it quitting/missing proving it's getting enough fuel?

It's also possible something is failing with temperature (even the ECU). Heating components with a hair dryer with engine running can sometimes bring out temp issues.

Know that if it is the ECU, ANYTHING weird is possible.

The ECU when it decides everything is ok (getting proper signals in it's power up sequence) is the thing that activates the fuel pump (driving ECU pin 74 low) and ASD relay (driving ECU pin 67 low). So if the ECU doesn't see everything is proper it won't activate these. If that is the case when you are forcing the ASD, you can see the incompatibility with that and you would need to figure out what is wrong in the ECU power up sequence (keeping in mind the ECU itself may even be the problem). Or just try a different ECU.

Thank you for all of this information. Yeah the pump and filter are brand new. Didn't even think of messing with the regulator at the time that's my fault. And That's one of the weird things, if I let it sit for 5-10 minutes idling, and go and rev it to like 5-6k rpms for a bit, it acts perfect. But then 1-5 minutes after actually shifting through gears driving it dies. I'm pretty confident she could idle all day and never die. Tonight I noticed if I slam on the gas just as it tries to die I can sometimes save it from fully dying so maybe your onto something with the relay. I'll get a replacement and see what happens. If that doesn't help I'll probably try an ECU just to see. Thank you so much for all of the help

 

[steve]

Even Steve who is a turbo expert knows ECU operation better than anyone so may know the NT too.

I don't and have tried to ignore 2GNT's but I guess I need to do an ASD write up.