9 month long problem - no start

[scorpion66]

I have spent over 9 months trying to figure out whats wrong with this car, and I'm at my wits end. One of my mechanic friends refuses to help anymore and just wants me to give up, and the other is agitated and thinks I should put it in a shop. I'm really close to doing that, but before I do, I wanted to run this by you all and see if anyone had any ideas.

I'll try to summarize with bullet notes at the bottom incase ya dont feel like reading a book.

Details:
1996 GS Eclipse, 128k miles, all stock except air intake and a msd ignition pack.
recently replaced timing belt and waterpump


Beginning:
Was driving along a highway and caught a glimpse of the battery light flickering a couple times on the dash, then it stopped. Car didn't seem to have any problems. I made a stop, shut the car down, and about 5 minutes later when I got back in the car and went to start, it seemed to fight a little bit. It did start after the second or third try, I didn't really think anything of it, except maybe that the battery was old and needed to be replaced, since there wasnt really any funny noises or anything.
I drove the car home, parked it and went inside. Half an hour later I came out to take the car to autozone to have the batt checked and possiby replaced if need be. Went to turn the car on, but no go. It would turn over again and again, but never actually start.
It didn't sound like it was cranking wildly or anything.
I figured I'd take the battery in with my other car and have it checked after a couple more tries. Found out the battery was shot. They said they didn't know how it was even turning the car over. I got a new batt, took it home, and just for the hell of it, popped it in and tried it one more time. Nothing. It turned over just like before but the car didnt start.
I did notice at this point that the check engine light did NOT come on with the rest of the dummy lights when I turn the key to the on position. I even tried to get a code from it by turning the key on and off. Nothing. I know the CEL was working the day before. I pulled the dash out real quick and checked the CEL bulb, it was fine, but just to be sure I swapped it with a bulb that was definately working. replaced everything and checked again. Still no CEL.
I popped the hood and checked the connections to the ECU, they all seemed fine.




After talking with a couple people over the span of months, this is what I have done so far:
Had the angel positioning sensor and coilpack tested - both ok. Replaced the ECU. Replaced the plugs and wires during all this after noticing a couple cracks in the old wires. (yes they are gapped correctly)




I had a friend come over at one point and give me a hand, he said he checked for spark and it was getting spark. He also said he checked for pressure in the fuel rail and it seemed fine.

I went ahead and put the new ecu in, and still nothing, not even a CEL. I went and checked for spark, I don't know if maybe I did it wrong, but I didn't see it sparking.
There was however pressure in the fuel rail.

I spoke with another friend and he said that while checking all those things is surely helpful, he doubts I will find the problem there. He basically said that because there is no start-up CEL that there is likely no communication with the ECU. He couldnt think of anything specifically that would prevent it other than the ECU not getting the power and grounds it needs.
He suggested checking the entire 12v system, especially any grounds near the ECU.
Thats a pretty large task though, especially for me not having any wiring diagrams. I really dont want to be blindly poking around in there and mess anything else up in the process.



Summary:
- 1996 GS Eclipse, 128k miles
- stock except air intake and a msd ignition pack
- recently replaced timing belt and waterpump
- new plugs and wires
- fuel pressure - ok
- Won't start, but turns over fine (not wildly spinning though)
- No Check Engine Light at startup
- No CEL codes
- Checked CEL bulb and swapped with working bulb - no effect
- Checked angel positioning sensor and coilpack - both checked out ok
- Replaced ECU - no effect
- Can not determine that there is spark (need help here)
- Believe that the problem resides in ECU wiring (need help here)


Aside from the suggestion to check the wiring, I am completely out of ideas. And I fear I may have to put the car into a shop, and pray that they don't take me for every penny I have. I can't stand putting a car in for work when I don't have an idea of what the problem is. Do any of you guys have any thoughts, suggestions, or even questions? Maybe theres something I missed?
Keep in mind I am not a mechanic, so be gentle Thanks for all your time.

 

[laserspeeddemon]

Check www.mitsubishigraveyard.com they will likely have a whole harness. If you're harness did indeed melt together as you said, then it would be obvious. You'd smell it. And something like that would smell for weeks. It's not the most likley source of failure. I have to go for now, but tomorrow I will search around for a diagram for the N/T 2g.

 

[Locke]

My assumption is that somewhere along the harness, the wires are melted together,

A common problem with 2gnt's.

There is a loom of wires running near the EGR tube, on the driver side of the engine, under the intake tube. The EGR tube get very hot, and often shorts wires together within the loom.

You will have to pull the loom off, and all the gobs of electrical tape to inspect. The wires sometimes are damaged without noticeable signs of damage to the loom or electrical tape itself.

 

[chrisfullwood]

Have you considered the ECU, if the ECU isnt showing the CEL, its not running its self diagnosic if im mistaken.

If the ECU is having problems, it would definately cause a no spark issue, depending on where the problem was. Leaking Caps are a big problem with DSM cars.
might be worth pulling it out, popping the cover off and taking a look. i read through the posts and didnt see any suggestions to find an ecu that matches yours and test it.
look very carefully at the board inside the ecu for any acid marks.
good luck. remember the old theory from sherlock holmes. after you have elliminated all logical answers, whatever remains, no matter how crazy it is.. must be the answer.

 

[scorpion66]

Summary:
- 1996 GS Eclipse, 128k miles
- stock except air intake and a msd ignition pack
- recently replaced timing belt and waterpump
- new plugs and wires
- fuel pressure - ok
- Won't start, but turns over fine (not wildly spinning though)
- No Check Engine Light at startup
- No CEL codes
- Checked CEL bulb and swapped with working bulb - no effect
- Checked angel positioning sensor and coilpack - both checked out ok
- Replaced ECU - no effect
- Can not determine that there is spark (need help here)
- Believe that the problem resides in ECU wiring (need help here)

I already thought of the ECU, in fact was on of the first thoughts. Was replaced, no effect.

 

[chrisfullwood]

you could always hook up a timing light to the plug wires one at a time and see if they are firing?
that will show you if the plug wire is getting the pulse for the spark.

my bad, i thought i read it... but i was at work at 5am.. till 9pm.

if you do not have the wiring harness diagram, let me know i might be able to get it for you.
my email is [email protected]
shoot me a mail and let me know so i can reply with it for you.

 

[scorpion66]

Thank goes out to chrisfullwood, the diagrams are great, I'm going through the harness a wire at a time. If its a wiring problem, I'll find it.

Thanks again chris

 

[laserspeeddemon]

Wow, have fun with that.

 

[scorpion66]

Ok, so I bought a shitload of electrical tape and loam, stripped the old loam off one section at a time and checked every single wire . 14 hours, a couple bloody knuckles, and 2 blackened arms later, I have finished pulling all the old loam and tape, checked each individual wire visually (some of those things are hard as hell to get to), and have found nothing wrong.

Nothing. Not one frayed, cut, or burnt wire, anywhere. On a positive note, now my car has some spiffy looking red loam/tape to match the red paint job. I guess by the time I'm done tracking down the problem, I'll have at least made the engine bay spotless. Since I already cleaned just about everything in there so I could see better and work easier. Judging by the pile of rags, I'd say it has been waiting to be cleaned like that for about 10 years or so

Matter of fact, I got so bored one day while trying to think of something else to check, I pulled the corroded exhaust guard off, wire brushed and sanded it with a fine grain pad. nice 'n shiny now. Just cant get the damn thing running


Another positive note a friend of mine is trying to get a buddy he knows that works at the Mitsu dealer's shop to come over take a look and plug a tester into the car and see if he can figure out anything. Hope he has better luck than me.

In the mean time, anyone else thought of anything new? I'm just kinda sitting around at the moment, cause I'm out of ideas.

 

[scorpion66]

I could just be "LOOKING" for something to be wrong here, but just in case...

For whatever reason, I decided to pull the little view "window" on the timing belt case and peek inside.
I have never done a timing belt myself, nor ever helped with one. I do know that the positioning of the gears, especially on an interferance engine like my 420a, is vital. I noticed a couple paint pen marks on the gears, but they were almost a half a turn from lining up with each other. I couldnt see any other marks through the tiny window.
My question in this bit has to do with how tight the timingbelt should be. I've always assumed that there should be no give in those belts when just pulling with one finger.
When I reached inside and gave a tug upwards, the belt stretched almost enough for me to have touched it to the top of the inside of the timingbelt housing.
Just how stiff are these things supposed to be? rock hard to the point theres no movement? or is a small bit of give ok?
Next question: If the new timing belt/tensioner/waterpump failed, and the belt did skip a tooth or two (praying to god the valves are ok if it did), would that keep the check engine light from coming on at startup? Since the CAS and CPS wouldnt be getting the correct signal?
Is skipping a couple teeth common at shutdowns/startups? (because remember, the car was running perfectly the last time it was driven)
One other thing I notced, the timing belt seemed to only be wide enough to cover about 1/2 of the gears and was all the way back against the back of the housing. Is that normal? or Should it be closer to 3/4 of the width of the gears or bigger? Should it be centered on the gears? And if so, is the fact that its off to one side indicitive that it may have slipped?


I really am not sure if this could be the source of my problem. I'm not even sure if I want it to be the problem or not. On one hand, its scary to think that I may have internal parts damaged, while on the other it would be great to have finally found the source of the problem.

Since the timing belt, tensioner and water pump were all replaced recently in a local shop (I would guess less than 2000 miles ago), if the timing belt skipped a tooth or more, whether theres internal damage or not, should I expect the shop to replace/repair any or all of the damage or the belt/tensioner? Or is this gonna be one of those things where I eat the whole cost?

 

[scorpion66]

I had said earlier that I was going to test all the fuses.

Just wanted to make a note that I did that. Checked all the connections as well.

Only found one burnt fuse and that was reverse lights (yes I check the fusebox under the dash and the one in the engine compartment) I replaced that while I was there.

I also noticed pieces of an old aftermarket security system, I went ahead and removed that just incase that was interfereing with something (and because I didnt have anything better to do).

Still nothing.

I have yet to find that ICM, even with pulling all the wires apart and searching. I must be missing something? Would it be under the engine? maybe inside the car under the dash or something?

 

[luv2rallye]

ICM = Ignition Control Module but DSM's don't call it that. 2g NT's have a Powertrain Control Module (PCM but some still call it ECU) in engine compartment driver's side which controls ignition - #21 in attached image post 2: http://www.dsmtuners.com/forums/showthread.php?t=100294. But you said you have a MSD ignition system which will be interconnected with the PCM and that is probably your problem with no spark. You also could try disconnecting the filter capacitor to see effect (in case it's bad).

 

[scorpion66]

Alright well, I'm 99% sure its NOT the ECU/PCM that is the problem, laser would agree with me there I'm sure. The odds of the ECU I had originally being bad AND the refurbished (and tested) ECU I just replaced it with just are way to slim to consider that a probable cause. I know someone who has gone through that refurbish company before on several occasions and had nothing but great stuff to say about them.

I was under the impression from posts here, that an ICM was in fact a totally seperate beast, not the ECU itself. But, moving along.

The MSD ignition pack has been tested, and works fine. The plugs and wires are brand new. The wires into it have also been tested, the ground and 12v tested fine but the signal wire never sent a signal during attempted startups.

Following that wire back to the ECU, it is clearly intact, with no defects or damage. Assuming the ECU(s) aren't the problem themselves. What would be the next object(s) in line that would be preventing the ECU from either powering up, or sending the signal to the coilpack?
Along with that, is there a reliable way to see if the ECU is actually powering up while in the car?

As for that filter capacitor, I checked my manual, the images you linked, and a couple pages of searches on google, and didnt turn up the location it can be found. I did note something said a capacitor can be found on the top left of the engine, but thats not entirely descriptive, and it would just make life easier knowing exactly where to look tomorrow at sunrise. Any idea where it is?

BTW laser, you've been aweful quiet, any thoughts?

 

[laserspeeddemon]

I'm gonna go look through a 2g manual.

**edit** It seems the haynes manual is too broad, doesn't help me at all.

 

[luv2rallye]

With battery disconnected did you test the continuity of the ignition wires with a multimeter (not just look at them) and also make sure there's no short to ground on them?

Here's a pic of the NT noise capacitor (from 99 NT factory manual). It usually is a metal can with 1 wire coming out of it connected to the power side of the coil primary.

 

[scorpion66]

Sorry I haven't replied, been busy with life I guess. Wifes birthday and all that.

Anyway, gonna go out and look for that noise capacitor.

And yes, I did test them with a multimeter with the batt disconnected.

 

[chrisfullwood]

as far as i know, there shouldnt be a whole lot of play in the timing belt.. i know sometimes there can be depending on where the tension is on the belt when it stops.

but you shouldnt be able to flex it more than 1/2" or so. if you can move it that much, i would definately check your tensioner, and it could easily jump 2 teeth.
my fiance drove my car to tulsa, ran fine on the way there, i tuned it a bit to make it run cooler.
she got to tulsa, stopped the car. went back on, and tap tap tap tap tap. but the car ran. and i realigned the cam gears. i still lost compression, because the tensioner was allowing the belt to flex, bleeding off the compression.
so definately check all the components of the timing. with as much flex as your talking about, there is definately something wrong there.

hope that helps, i can probably get the information on aligning your cam gears if needed.
later
Chris

 

[laserspeeddemon]

Honestly I'm at my wits end. Sorry guy, I can't do much else without the car in front of me.

 

[scorpion66]

Just a follow up on this, for all those of you who helped with ideas and thoughts.


The car is running now, running perfect. Better than before.

The culprit was apparently the ECU after all. I had tested 3 of the things in it (the original, a refurbished one I purchased from a company whos name I won't release just yet until they decide if theyre refunding my money after sending me an incorrect computer, and a known running one that my mechanic had in one of his used cars on his lot.)
As soon as he plugged in the good ecu the car fired up. He pulled the first ecu apart and found that those infamous capacitors killed the original ecu. He also tested the car with the refurbished one in, and found out it wasn't running with that because it wasn't the right part. Wrong part numbers, for the wrong year, model, and also for a auto instead of a 5spd.

Total shop work costs:
labor $125
diagnosis $100
ECU $450
------------
$625

Then theres also the refurbished ecu that I ordered 6 months ago or so, that turned out to be the wrong computer (their fault and I have the receipt and order form to prove it) which cost $325.
All of the odds and ends I bought at the parts store to work on the car in hopes of finding the problem on my own over the months total roughly around $80

So grandtotal $1030
Unless I can convince that company to refund my money and take their part back.



Anyway, thanks for all the help over the whole process guys.

 

[misphit]

What a brutal thread, scorpion... I once had a non-dsm that posed a similar no-start problem lasting almost three months (turned out to be the coil)... I feel your pain.

 

[laserspeeddemon]

Well. Glad you could figure it out... Sorry I wasn't much help.

 

[laserspeeddemon]

Well. Glad you could figure it out... Sorry I wasn't much help.