The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Fuel Injector Clinic
Please Support STM Tuned

2G Car still won’t start

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

98gstJames

Proven Member
215
41
May 25, 2022
WInchester, Virginia
Alright so I’ve been trying to get my car to start and I’ve done a lot, what does it sound like my car is missing now? I’ll add a video, I’m just lost

You must be logged in to view this image or video.
 
James that video of the sparks doesn't look right. The pattern should be:

1 & 4 spark (cyl. #1 fires)
2 & 3 spark (cyl. #3 fires)
1 & 4 spark (cyl. #4 fires)
2 & 3 spark (cyl. #2 fires)
1 & 4 spark (cyl. #1 fires again ...)

1-4 and 2-3 plug pairs alternate with one pair firing and then the other. Which cylinder actually fires depends on which one just compressed a fuel/air charge.

With the plugs laid out like that you should see one pair then the other, every time. But it looks to me like your 1-4 pair mostly fires only half as often as your 2-3 pair. If I'm seeing that right, it's wrong -- either cyl 1 or cyl 4 is almost never getting fired.

Paul if you didn't look at that video would you double check me?

The stethoscope info is good. It doesn't exactly match the spark pattern but maybe there are two problems. That's pretty common on old & neglected cars. #4 'occasionally' tic'ing combined with missing almost half the sparks for 1-4 pair suggests an ECU problem I think? Plus maybe a separate problem with #3 injector or its driver transistor in the ECU.

I would definitely look inside that ECU. Also be sure it's the right one for your car -- I have seen screwy stuff when a car had the wrong ECU for the engine or car model.

Paul does this make sense to you? You've got a lot more experience with these cars than I have.

James you've got some excellent data here. Man it's no wonder this car never ran decently.
 
Last edited:
Now you might be getting somewhere. The fast tick tick tick is what you should hear. The ecu light going off is correct. Even so you may still have an ecu problem. Have you looked inside? Sometimes there is visible damage and corrosion and a likely culprit for mystery problems.
Good call on repairing the injector wires. You could still use a noid light set to see how they are firing but it sounds like repairs need to be made anyway. Wouldn't hurt and it's cheap. Keep plugging away.
Yes I have looked inside the ecu for damage, that’s what I was looking into when my car wouldn’t even click at the starter when I turned the key(it was the starter), I opened the ecu to look for burns or corrosion and everything looked perfect inside there. I’m going today to fiddle with the injector wiring, I’m gonna look down the harness more and make sure they’re aren’t any deeper cracks or anything. Btw thanks for all the help, I wouldn’t be nearly as far as I am now without the help.

James that video of the sparks doesn't look right. The pattern should be:

1 & 4 spark (cyl. #1 fires)
2 & 3 spark (cyl. #3 fires)
1 & 4 spark (cyl. #4 fires)
2 & 3 spark (cyl. #2 fires)
1 & 4 spark (cyl. #1 fires again ...)

1-4 and 2-3 plug pairs alternate with one pair firing and then the other. Which cylinder actually fires depends on which one just compressed a fuel/air charge.

With the plugs laid out like that you should see one pair then the other, every time. But it looks to me like your 1-4 pair mostly fires only half as often as your 2-3 pair. If I'm seeing that right, it's wrong -- either cyl 1 or cyl 4 is almost never getting fired.

Paul if you didn't look at that video would you double check me?

The stethoscope info is good. It doesn't exactly match the spark pattern but maybe there are two problems. That's pretty common on old & neglected cars. #4 'occasionally' tic'ing combined with missing almost half the sparks for 1-4 pair suggests an ECU problem I think? Plus maybe a separate problem with #3 injector or its driver transistor in the ECU.

I would definitely look inside that ECU. Also be sure it's the right one for your car -- I have seen screwy stuff when a car had the wrong ECU for the engine or car model.

Paul does this make sense to you? You've got a lot more experience with these cars than I have.

James you've got some excellent data here. Man it's no wonder this car never ran decently.
I looked in the ecu a few weeks back and everything looked fine inside, it has a Mitsubishi logo on it and doesn’t look any different than the ones I’ve looked up, but how should I go about fixing this incorrect firing order?
 
James that video of the sparks doesn't look right. The pattern should be:

1 & 4 spark (cyl. #1 fires)
2 & 3 spark (cyl. #3 fires)
1 & 4 spark (cyl. #4 fires)
2 & 3 spark (cyl. #2 fires)
1 & 4 spark (cyl. #1 fires again ...)

1-4 and 2-3 plug pairs alternate with one pair firing and then the other. Which cylinder actually fires depends on which one just compressed a fuel/air charge.

With the plugs laid out like that you should see one pair then the other, every time. But it looks to me like your 1-4 pair mostly fires only half as often as your 2-3 pair. If I'm seeing that right, it's wrong -- either cyl 1 or cyl 4 is almost never getting fired.

Paul if you didn't look at that video would you double check me?

The stethoscope info is good. It doesn't exactly match the spark pattern but maybe there are two problems. That's pretty common on old & neglected cars. #4 'occasionally' tic'ing combined with missing almost half the sparks for 1-4 pair suggests an ECU problem I think? Plus maybe a separate problem with #3 injector or its driver transistor in the ECU.

I would definitely look inside that ECU. Also be sure it's the right one for your car -- I have seen screwy stuff when a car had the wrong ECU for the engine or car model.

Paul does this make sense to you? You've got a lot more experience with these cars than I have.

James you've got some excellent data here. Man it's no wonder this car never ran decently.
I didn't see it. The rate of fire would be rpm dependent while turning over. It looked normal to me. 14 and 23 taking turns.
 
Nope, That's not an ECU. That is the Cruise Control Computer.

The ECU for a 1998 GST is a MD346675, one of the black plastic ECU's like this. Given the mods on the car it's possible there is a different ECU in the car from the original.

You must be logged in to view this image or video.
 
Last edited:
I didn't see it. The rate of fire would be rpm dependent while turning over. It looked normal to me. 14 and 23 taking turns.
Right about RPM dependent of course but they don't seem to be taking fair turns. In that clip I count 11 sparks for #2 (and same for #3) but only 6 for #1 and #4. This cannot be right.

However: The question is, am I wrong about my counting? Man, I can't get anything but 11 and 6, but ...

Another piece of evidence might be that you couldn't hear anything from the #4 injector. If the ECU rarely decides to fire the #4 cylinder maybe it loses both that injector pulse and that spark?

I'm kind of at a dead end for advice: If the number of sparks isn't the same the odds are good that it's an ECU problem and the best way to test that is a substitute ECU -- you can do that without removing the old one. You just pull the four plugs (like all the others on these cars they have press to release latches), flip them back, and plug them in to your test ECU leaving the old one in place while you are testing.

Maybe do your own count of sparks on each plug in that video and decide whether to try a substitute ECU?

Edit: Another idea would be to shoot that video again but make it maybe 2 or 3 times longer. About the only thing I can think of that would fake the count would be some of the sparks happening between frames of the video; I don't know enough about that tech to say how likely that is but another and longer video should show it as the reason for the missing sparks if it is.

Definitely an interesting problem.
 
Last edited:
Well the op has some work to do as it is clear he did not check ECU. the car also has some sort if td05 based turbo. And fmic etc. I'm betting it has fuel mods. If it doesn't previous owner might have caused undisclosed damage.
 
Right about RPM dependent of course but they don't seem to be taking fair turns. In that clip I count 11 sparks for #2 (and same for #3) but only 6 for #1 and #4. This cannot be right.

However: The question is, am I wrong about my counting? Man, I can't get anything but 11 and 6, but ...

Another piece of evidence might be that you couldn't hear anything from the #4 injector. If the ECU rarely decides to fire the #4 cylinder maybe it loses both that injector pulse and that spark?

I'm kind of at a dead end for advice: If the number of sparks isn't the same the odds are good that it's an ECU problem and the best way to test that is a substitute ECU -- you can do that without removing the old one. You just pull the four plugs (like all the others on these cars they have press to release latches), flip them back, and plug them in to your test ECU leaving the old one in place while you are testing.

Maybe do your own count of sparks on each plug in that video and decide whether to try a substitute ECU?

Edit: Another idea would be to shoot that video again but make it maybe 2 or 3 times longer. About the only thing I can think of that would fake the count would be some of the sparks happening between frames of the video; I don't know enough about that tech to say how likely that is but another and longer video should show it as the reason for the missing sparks if it is.

Definitely an interesting problem.
I didn’t notice it the first time but after reading the debate I went back and rewatched. 1/4 are for sure skipping some of their turns.
 
Without knowing the camera frame rate I'm not sure I would trust counting the recorded sparks.
I agree with you. That would be a perfectly innocent explanation -- just an artifact of using a video camera to photograph very brief events. My suggestion of a 2x - 3x longer video was in the belief that if it is a meaningless oddity we'll -- probably -- be able to tell.
 
I'm with Steve could be video but you're right. The video does show an issue. We just don't know if its real.
I see yet another possible issue with the video rather than the car. To help figure this out, James, if you'll do another video say 2x the length and instead of arranging the plugs 1-2-3-4 from top to bottom, lay them out 2-1-4-3 ?

General use consumer video 'cheats.' It provides more detail where your eye is most likely looking (the center of a picture) and less in other areas. One way to do this is to effectively lower the frame rate away from the center: For most action pictures this would just mean a bit of jumpiness or smearing near the edges but for video of events (sparks ...) it might mean some completely disappear.

Reversing the pairs between the ends and the center and shooting a longer clip should clear this up. This is either a powerful clue or ... nothing at all.
 
I see yet another possible issue with the video rather than the car. To help figure this out, James, if you'll do another video say 2x the length and instead of arranging the plugs 1-2-3-4 from top to bottom, lay them out 2-1-4-3 ?

General use consumer video 'cheats.' It provides more detail where your eye is most likely looking (the center of a picture) and less in other areas. One way to do this is to effectively lower the frame rate away from the center: For most action pictures this would just mean a bit of jumpiness or smearing near the edges but for video of events (sparks ...) it might mean some completely disappear.

Reversing the pairs between the ends and the center and shooting a longer clip should clear this up. This is either a powerful clue or ... nothing at all.
Alright I’ll be taking the plugs out and recording a loner video, the last one shouldn’t have lost and frames it was just slowed down not recorded in slow motion, not sure if that affects it or not, but I’ll also be tracking down this ecu now that I know what I thought was the ecu is not. What kind of undisclosed damage could there be? I know it’s the stock fuel pump in the car bc I had to pull the old fuel out so the pump had to be pulled a little bit to do it.
 
Can you explain this fuel pump statement? I'm confused. Pulled a little bit???

On ecu just pull it and tell us what you have. It's either a black box or a gold one. What you find inside, well we will see.
 
Can you explain this fuel pump statement? I'm confused. Pulled a little bit???

On ecu just pull it and tell us what you have. It's either a black box or a gold one. What you find inside, well we will see.
There was old gas in the tank and you can’t put a siphon hose down the filler neck so I loosened the bolts on top of the fuel pump and popped it off just enough to put the hose down into the tank so I could put fresh fuel in it.
 
There was old gas in the tank and you can’t put a siphon hose down the filler neck so I loosened the bolts on top of the fuel pump and popped it off just enough to put the hose down into the tank so I could put fresh fuel in it.
So how did this tell you its a stock fuel pump? Many upgrades don't look any different than stock. Either way I'm still wondering if there are ecu and injector mods.
 
So how did this tell you its a stock fuel pump? Many upgrades don't look any different than stock. Either way I'm still wondering if there are ecu and injector mods.
When I looked at the fuel pump it didn’t look like any of the aftermarket looking ones just look like the gold factory one and I had to take the fuel rail off to replace a lower insulator on one of my fuel injectors and they look like the factory ones but like you just said they don’t have to look different to be different but I assume the injectors and pump would look a little different?
 
When I looked at the fuel pump it didn’t look like any of the aftermarket looking ones just look like the gold factory one and I had to take the fuel rail off to replace a lower insulator on one of my fuel injectors and they look like the factory ones but like you just said they don’t have to look different to be different but I assume the injectors and pump would look a little different?
That the operative word. Assume. Untrue on both accounts.
It may not matter right now but you're going to need to know eventually. Meanwhile what about the ecu??
 
Alright I’ll be taking the plugs out and recording a loner video, the last one shouldn’t have lost and frames it was just slowed down not recorded in slow motion, not sure if that affects it or not, but I’ll also be tracking down this ecu now that I know what I thought was the ecu is not. What kind of undisclosed damage could there be? I know it’s the stock fuel pump in the car bc I had to pull the old fuel out so the pump had to be pulled a little bit to do it.
A longer video just like that one with the plugs in order 2-1-4-3 so we can see whether 1 and 4 are really getting half as many sparks as the other two or does it just look that way because of how the camera works.

The issue isn't 'lost frames' but (a) what happens between frames -- is there a dead spot where a spark might not show up? and (b) is the frame rate really the same everywhere or is the little computer in the camera looking more often at the center of the picture than the rest of it?

If you're watching the plugs while the engine is cranked you might be able to tell just looking: In the video 2-3 are quite clearly sparking twice for every one spark on 1-4.

This is what it's always like diagnosing weird problems. You cannot move ahead until you're certain about your data. And this definitely scores as a weird problem.

I’ll also be tracking down this ecu ... What kind of undisclosed damage could there be?
These ECUs -- particularly the earlier ones -- got really bad electrolytic capacitors. Often they fail, leak corrosive stuff on the circuit board, and cause other stuff to fail. You can see these failures just by taking off the box and looking closely. Sometimes they can be repaired -- there are businesses that specialize in that -- sometimes not. (I'm 0 for 2) If your '98 car has the ECU that came in it those failures are less likely I think -- Mitsu had by then starting using better caps.

But there's lots we don't know: Many of these cars will run with an older/the wrong ECU. Maybe the prior owner made a swap. I got a car like that (a '94 Expo LRV) and it took a while to figure it out. Even if your car runs okay with (say) a '95 ECU you'd have the older unreliable caps and one or more of them might have failed.

However: Even if a visual inspection says the ECU circuit board looks okay there are other possible failures -- the CPU chip itself can die. Or the transistors that actually give the CPU commands to the outside world can fail. These failures are less notorious but far from rare.

I wouldn't rush to pull the ECU out right now. Do get the number -- the last four digits are on the front and the whole part number is on the top -- and when you open things up to where you can see it, post that number here. Someone will be able to tell you if it's a box that will work right in your car or not.

Because you can substitute another ECU just by pulling the plugs out of the old one and shoving them in the new,, and a nice clean circuit board isn't proof that the thing works, I wouldn't now pull it out just to inspect it.

If you substitute an ECU and that fixes things, yeah, pull and replace. Or if it turns out the ECU isn't the problem then your list of stuff to do later should include 'inspect ECU board' because if there's corrosion on it, it will fail and you want it cleaned and repaired before that happens if possible.

I am not as worried about mods to your car as others here. Yes there are real issues there but when mods keep a car from starting at all they do it just the way failed parts do -- no spark, no injection in a couple cylinders, etc. Keep on troubleshooting. When the car starts and runs but not very well then you're going to have go back and figure out exactly what's been done and whether it was done correctly or not, right along with looking for smaller failures -- bad sensors and stuff like that.

Our goal right now is to get it to start and run.
 
These ECUs -- particularly the earlier ones -- got really bad electrolytic capacitors. Often they fail, leak corrosive stuff on the circuit board, and cause other stuff to fail. You can see these failures just by taking off the box and looking closely. Sometimes they can be repaired -- there are businesses that specialize in that -- sometimes not. (I'm 0 for 2) If your '98 car has the ECU that came in it those failures are less likely I think -- Mitsu had by then starting using better caps.

But there's lots we don't know: Many of these cars will run with an older/the wrong ECU. Maybe the prior owner made a swap. I got a car like that (a '94 Expo LRV) and it took a while to figure it out. Even if your car runs okay with (say) a '95 ECU you'd have the older unreliable caps and one or more of them might have failed.

However: Even if a visual inspection says the ECU circuit board looks okay there are other possible failures -- the CPU chip itself can die. Or the transistors that actually give the CPU commands to the outside world can fail. These failures are less notorious but far from rare.

I wouldn't rush to pull the ECU out right now. Do get the number -- the last four digits are on the front and the whole part number is on the top -- and when you open things up to where you can see it, post that number here. Someone will be able to tell you if it's a box that will work right in your car or not.

Because you can substitute another ECU just by pulling the plugs out of the old one and shoving them in the new,, and a nice clean circuit board isn't proof that the thing works, I wouldn't now pull it out just to inspect it.

If you substitute an ECU and that fixes things, yeah, pull and replace. Or if it turns out the ECU isn't the problem then your list of stuff to do later should include 'inspect ECU board' because if there's corrosion on it, it will fail and you want it cleaned and repaired before that happens if possible.

I am not as worried about mods to your car as others here. Yes there are real issues there but when mods keep a car from starting at all they do it just the way failed parts do -- no spark, no injection in a couple cylinders, etc. Keep on troubleshooting. When the car starts and runs but not very well then you're going to have go back and figure out exactly what's been done and whether it was done correctly or not, right along with looking for smaller failures -- bad sensors and stuff like that.

Our goal right now is to get it to start and run.
Alright well I didn’t see these replies since my last one and I went ahead and pulled the ecu and took it apart to look inside, I didn’t see anything it here’s what it looks like.

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.
 
Surface mount technology......very hard to diagnose with the eye.
 
Surface mount technology......very hard to diagnose with the eye.
Agreed but there are two places offhand that grab my attention.

On the bottom the R701 resistor looks unusual.

You must be logged in to view this image or video.


On the top there is a circuit trace in an area of silkscreening that doesn't look right.

You must be logged in to view this image or video.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top