car runs. For two seconds Need some help

[cantrun10s]

Several questions here but I'll try to keep it simple.

Background information - Fresh rebuild of a completely stock motor. New Valves, guides, seats, seals, pistons, rods, pins, stock turbo, stock piping, stock vacuum routing. This is a completely unmodified car, totally rebuilt, twice. First rebuild suffered an incident with the timing belt, and lead to the second rebuild. Computer has been replaced (thanks Steve).

At this point i'm not interested in modifications untill the motor has at least 5k miles on it, but it wont even run as intended.

Symptoms - With the new ECM installed it goes as follows
Turn key on - boost gauge jumps to zero, seat belts come out, everything working as intended. Turn key to start position - engine cranks. And cranks and cranks and cranks. No fire. Return key to run position - fuel pump turns on. Will not turn on in crank position, only runs for two seconds, and then shuts off (which i understand is normal). Turn key to Start position - car runs. For two seconds, then sputters out and dies. After this it's always the same. Pops right off, runs for two seconds, dies.

I'm pretty mildly tempered, but with the amount of money I've got into this motor it really pissed me off. So heres the rest of it. You can hold the throttle to the floor after the initial crank, and it will start, and hit about 5k on the tach, and then it's very similar to fuel cut, but it cuts spark and fuel at the same time.

Hot wire the fuel pump - car starts, runs for two seconds, cuts spark.

Any ideas?

Also - car is a 91 GS-T

The ECM i got from steve runs my 90 GSX like a dream, but on the same note, the computer out of my 94 non turbo talon makes the 91 run.

I believe the problem would be in whatever signal wires the ECM uses to control fuel and spark, or even what it uses to determine if the car is even running.

 

[talon_frost]

its your ecu if you got a ecu from a different year you need to make sure its turbo and the same drive line suck as awd and 5 speed....then if you got the ecu from a different year like 91-94 are the same but a 90 is different you need to swap 2 wires to make the car run . the pins are 6 and 14 swap those and see how it runs....

 

[cantrun10s]

Computer I have is a Turbo ECM, EPROM, Socketed, And was origianally in a car identical to the one it was pulled from. Same model number, same series number. Before i swap any wires in the ECM I'll be hooking it to the Solus and pulling readouts from every sensor the car monitiors, and I'll just have to take it from there.

But you did peak my interest. What exactly are pins 6 and 14?

{EDIT} - Thanks to DSMlink again. They provide a VERY nice Pinout diagram. I will be looking into this right now.

 

[RayBan GST]

change your crank sensor and let me know

i bet its the problem


good luck

 

[cantrun10s]

CAS new by Beck Arnley. ECM is reading it. Will swap out with one off my 90 and see if anything changes.

 

[Daaan]

The exact same thing happens to my car but only if its been running for a while and the engine is hot. Sounds as if the car runs on 3 cyls and dies right?? The shop i know said to check the power transistor, coil, and crank sensor. Rayban gst might be right. Try a new crank sensor.

 

[92awddsm]

What is the ecu #?

 

[cantrun10s]

MD166260

6260 on the side

are the eprom ecm numbers.

 

[cantrun10s]

Started removing the body harness today. Anybody have any more suggestions? Have found no breaks in wires, swapping pins didn't do anything. CAS has been swapped with two known good ones, no improvement.

 

[srumel]

Were you able to pull any codes?

 

[cantrun10s]

No it never threw codes that I could read. I'm using a Snap-On Solus to watch readouts and pull codes, so I was never really able to get a good data stream on it, as it only runs for two seconds, was only able to verify that the sensors were at least sending...... something

 

[Anarchology]

Did you make sure you ignition timed the car with a timing light? I know that when I had my engine setup and newly installed, I had to time it properly with the CAS. You ground out the 'timing' wire (whatever it is known as), while setting the timing by advance/retard timing using the CAS to 5* BTDC. Remove the grounded wire that tells the ECU you are timing, and it should read around 8* BTDC and then advance more with revving from what I remember.

I might just be explaining something that you have already done. If so, then that ECU isn't getting the right signals.

 

[steve]

Symptoms - With the new ECM installed it goes as follows
Turn key on - boost gauge jumps to zero, seat belts come out, everything working as intended.
Turn key to start position - engine cranks. And cranks and cranks and cranks. No fire. Return key to run position - fuel pump turns on. Will not turn on in crank position, only runs for two seconds, and then shuts off (which i understand is normal).
Turn key to Start position - car runs. For two seconds, then sputters out and dies. After this it's always the same. Pops right off, runs for two seconds, dies.

So here's the rest of it. You can hold the throttle to the floor after the initial crank, and it will start, and hit about 5k on the tach, and then it's very similar to fuel cut, but it cuts spark and fuel at the same time.

Hot wire the fuel pump - car starts, runs for two seconds, cuts spark.

The ECM i got from steve runs my 90 GSX like a dream, but on the same note, the computer out of my 94 non turbo talon makes the 91 run.

Started removing the body harness today.

Since it seems your chasing your tail, lets go back. The engine runs without any of these issues when you use a NT ECU? If that's true why are you messing with the harness?

Do you know for sure that it cuts spark and fuel at the same time or are you guessing.

Have you verified the injector size and checked for vacuum leaks?

What's the current status and nature of the problem?

 

[cantrun10s]

I'm messing with the harness because the ECU i got does run my 90 very well. I'm aware that the difference between the 90 and the 91 ECU is two wires, but I have swapped those wires and got no improvement. I'm hoping to find something in the body harness, as I've already pulled and ohm'd the engine harness. I really cant explain why the 94 N/T ECU runs the car, as the only difference should be pins 6 and 14, and I have tried every possible concoction of ECM, wiring pairs, and other things.

I am for sure that it cuts spark. Haven't had that logged or anything but have had friend start it while I'm using an inductive probe to check things and all spark dies at the same time it cuts fuel. Injectors are stock sized, no vacuum leaks, boost tested to 20psi, at which point it starts blowing past my BoV.

Currently I'm considering contacting Painless to see if they do whole car kits. I've started to ohm out the grounds for the ECM, but haven't gotten into signal lines yet. Was trying to find a Pinout of the firewall connector.

I was also concidering pulling the motor and making this a winter build project, but i'd really like to at least put a couple hundred miles on the motor first.

 

[cantrun10s]

Did you make sure you ignition timed the car with a timing light? I know that when I had my engine setup and newly installed, I had to time it properly with the CAS. You ground out the 'timing' wire (whatever it is known as), while setting the timing by advance/retard timing using the CAS to 5* BTDC. Remove the grounded wire that tells the ECU you are timing, and it should read around 8* BTDC and then advance more with revving from what I remember.

I might just be explaining something that you have already done. If so, then that ECU isn't getting the right signals.

Well... Yes and no. I'm positive the timing belt is set correctly, and the CaS is at least not 180 off, but it could be a couple degrees off, as I'm not able to get the car to run long enough to get it perfectly set, but i'll guarantee it to be within 2-3 degrees of base time.

 

[steve]

I'm messing with the harness because the ECU i got does run my 90 very well. I'm aware that the difference between the 90 and the 91 ECU is two wires, but I have swapped those wires and got no improvement. I'm hoping to find something in the body harness, as I've already pulled and ohm'd the engine harness. I really cant explain why the 94 N/T ECU runs the car, as the only difference should be pins 6 and 14, and I have tried every possible concoction of ECM, wiring pairs, and other things.

The fact the the NT ECU works tells me that the basic wiring is functional. You have a 91+ ECU in a 91+ car so you don't have to change the pins on that harness. In your 90 the wires (6 and 14) would need to be swapped from what they originally were from the factory.

Running the NT ECU is going to give you a ton more fuel than the Turbo ECU. If your injectors are wrong size, wrong base fuel pressure, or a bad MAF your not going to get the right amount of fuel to keep the car running.

I am for sure that it cuts spark. Haven't had that logged or anything but have had friend start it while I'm using an inductive probe to check things and all spark dies at the same time it cuts fuel. Injectors are stock sized, no vacuum leaks, boost tested to 20psi, at which point it starts blowing past my BoV.

I would expect both to shut down at the same time when it dies. I think your seeing the result of it stalling not the cause.

If the car will run long enough with the NT ECU to get to operating temp there is no reason you can't use it to set the basic timing adjustment. That won't change between ECU's.

 

[cantrun10s]

I'm sure the N/T ECU would run it long enough to get to temp, I've just been afraid to let it run that long with that ECU. I was curious as to if I could get away with it or not, I'd like to not wash the rings in this one.

Also, I did not have to swap any wiring in the 90. Plugged it in, hit the key, took it around the block. Would the SAFC-II have anything to do with that?

 

[steve]

Also, I did not have to swap any wiring in the 90. Plugged it in, hit the key, took it around the block. Would the SAFC-II have anything to do with that?

No, the SAFC has nothing to do with it.
The pins we are talking about are used for the Idle Position Switch and the MAF reset line.

For a 91 ECU the green wire from the IPS should go to pin 14 on the ECU and the Green with White stripe MAF reset goes from pin 6 to pin 1 on the MAF. For a 90 ECU those are reversed.

You car might idle ok without the IPS functioning or the pins might already be swapped.
If your car idle higher than the stock 750 RPM and you haven't had a chip burned then the Idle circuit isn't working.

The NT ECU doesn't use the MAF reset line, the FPS, BCS, Boost Gauge, or Knock Sensor. These shouldn't have a work/no work interaction with the ECU. That's whey I'm leaning in the direction of fuel.

 

[cantrun10s]

I'm positive the injectors in the 91 are stock, but the rest of the fuel system is also stock, and original. I've replaced the filter, but that's it. So either way a pump / fpr isn't going to hurt me. The only other set of stock injectors I have came from my 90 GSX, which to my knowledge, are not the same size as a 91 GST. Or at the very least I know they run a different base pressure.


As much as I hate throwing parts at this thing without knowing what's wrong, here's my new plan of attack. I'll order a 255lph and an adjustable FPR tonight. I'll actually rewire the fuel pump the right way when I get it in, instead of my alligator clip 20 foot test lead...

Would anybody recommend trying the injectors out of the 90? Again I don't remember what the stock flow rate out of the 90 was, but I'm fairly certain it was different.

I'll do a cyl balance test on the injectors tonight, and NOID test the harness side.

I never intended to keep this car stock anyway, just wanted a small break in before I modded it.

Also 90 Idles closer to 900(with both ECM's), and I've never had a chip burned.

I'm positive the injectors in the 91 are stock, but the rest of the fuel system is also stock, and original. I've replaced the filter, but that's it. So either way a pump / fpr isn't going to hurt me.

And as far as the ISC circut goes, if that was my problem couldn't i keep the car running with the throttle? I'm also leaning tword fuel as being a problem, but I'm also not ruling out my MAF, and have a friend bringing over a different one tommorow so I can see if that changes anything. Only other MAF i have is out of the N/T, 90 runs MAF-T with a 3" GM, and the original is long gone. I'm basically sure that whatever is wrong with this car, it has the ability to cut fuel and spark at the same time. Ruling out the Cam sensor and the ECM, what else could cause the ECM to shut the car down? Faulty MAF signal? coolant temp below what paramater?

I've so far verified that the MAF is sending a signal, but have only been able to read it in terms of voltage, not hertz. Weither this is a result of my limitiations with the solus or the solus's limitations with DSM I have yet to figure out.

 

[steve]

The difference in injectors is between N/T, auto, and manual. A 90 should have the same injectors as the same style 91-94 car. If both cars are manuals, they should both have the same blue top 450cc injectors and say so in the side of the molded blue plastic body.

 

[cantrun10s]

Fuel system = the problem. Installed the 255, car runs. Not sure if it was a pressure issue or the pump, as i changed the pump, FPR, and rewired all at once, but one of those things was the issue. Did not change injectors.

 

[92awddsm]

Fuel system = the problem. Installed the 255, car runs. Not sure if it was a pressure issue or the pump, as i changed the pump, FPR, and rewired all at once, but one of those things was the issue. Did not change injectors.

How did the o-ring look on the pump outlet? cocked o-ring or cracked locater is the #1 cause of issues after a fuel pump install. Sounds like someone already swapped the pump and f'ed up the install which was causing the lean issues you were seeing.

 

[GSeX02]

def sounds like an ecu issue

 

[cantrun10s]

How did the o-ring look on the pump outlet? cocked o-ring or cracked locater is the #1 cause of issues after a fuel pump install. Sounds like someone already swapped the pump and f'ed up the install which was causing the lean issues you were seeing.

The o-ring was dry, and cracked. I dont know about swapped, but I'm certain that the tank had been out before. For what reason I'm not sure. The pump was a stock mitsu pump.

 

[steve]

Glad to hear you resolved it. Your fuel pressure must have been in the toilet.