Frustrating Issue - Car shuts off randomly, runs for no more than 10 seconds

[ktm]

The problem showed up two weeks ago when driving back from the SoCalROC Show-n-Shine Toy Drive. The show was 45 miles from my house. I've never experienced this issue before this time.

 

About 8 miles from my house the car misfired/sputtered/whatever once. It did it again about 6 miles from the house and again at around 4 miles. When I exited the freeway and came to a stop at the light at the bottom of the ramp, the car had shut off during decel. I managed to get it restarted only for it to die about 600 feet down the road. I coasted into a parking lot and attempted to restart the car. It would not run for more than 5 to 10 seconds at a time for a period of 5 minutes. I thought I may have run out of gas (gas sender unit has been acting up) and once I managed to get the car running one more time I managed to drive it 300 feet to a gas station connected to the parking lot.

 

I filled it up with gas and prayed that I could make it the 1 mile to my house. It ran without a problem at the stop light and all the way to the last major intersection before my house at which point it died again. I only had a 1/2-mile to go to my house and it died 3 times on the way to the house.

 

Once in my driveway it would only run for 10 seconds or less and then die. Not a sputtering shutdown but as though the car was turned off. I managed to keep it running for around 30 seconds by reving to around 2000 rpms, but even then it shut off in a similar manner.

 

I replaced the fuel pump with a brand new unit and installed two new filters (pre and post pump). It is still doing the same thing. I suspect it could be a lose connection at the ignition switch, a lose connection on the ignition circuit, or my fuel pressure regulator. However, when energizing the pump the fuel pressure reads 43 psi.

 

 

There are times when I cannot get the car to turn over at all.

I am asking for other possible causes and to confirm if a faulty FPR could cause these symptoms.

[Xnke]

I'd check your ignition switch for faults. If you need a new switch let me know, I have three or four around here somewhere if I can find them.

 

Are you using the factory injection system, or megasquirt, or what?

 

I'd check your igniton pickup connections, and your coil. Just start at one point, and since it acts like an electrical problem, treat it like one.

[Leon]

Really sounds like an ignition problem to me. I don't know what car/engine you have (L28ET?) but it sounds like it could be the black ignition box if you are using a ZX dizzy.

[ktm]

I am running a Wolf V500 EMS with LS-1 coil-on-plugs. I use the 1983 turbo distributor with a reference hole to #1 cylinder drilled next to the optical slot for #1. The ignition is setup for full sequential ignition and injection. I am not seeing errors on the EMS software screen (synch or noise).

 

It could be a loose ground for the coils, loose power for the coils, loose ignition circuit ground, bad ignition switch, etc. One odd thing I recalled last night was when the car shut off I heard the fuel pump key-on again. I could have imagined it, but if so then that could only occur if there was power interruption to the EMS since the fuel pump is set to run for 5 seconds after powering.

 

The symptoms also mirror a bad fuel pump which is why I replaced the pump and filters.

Edited December 12, 2010 by ktm

[zero2sixtyin3]

I am thinking it would be best if you could provide a little more info on what you are running. i mean what kind of car and setup are we trying to help you diagnose a problem on? as it stands it could be something like a bad MAF sensor or bad grounds of some sort. when it doesn't start at all does it seem like a lack of fuel or spark? seems to me everyone is on the right track but nobody can really offer more help with so little info aside from your ems...

[ktm]

1972 240z with a L28ET

 

Wolf V500 EMS

LS-1 COPs and triggered sequentially (not batch fired)

1983 turbo distributor with a reference hole to #1 cylinder drilled next to the optical slot for #1.

No MAF. Fuel and spark table setup using RPM and MAP.

TPS is only used for over-run fuel cut.

IAT and CTS only used for air temp and coolant temp compensation.

Walbro GSL392 pump

Aeromotive -10 micron filter post pump

Aeromotive -100 micron filter pre-pump

Aeromotive A1000 FPR

Siemens 757 cc/min injectors triggered sequentially (not batch fired)

 

Engine harness is all "new" wiring as of 4 years ago when I first wired up the EMS. New wires to the fuel pump were run at that time as well.

 

The car has never exhibited these symptoms before.

Edited December 12, 2010 by ktm

[rayaapp2]

Your going to have to bust out a multimeter or an O-scope.

Check the obvious. Fuel pump power and ground and watch for any interruptions that would lead you to further diagnose. IF its not there when the fault occurs move on to ignition. Verify you have basics here. Pull a COP off and load it with a spark plug. Ground the spark plug! OR you will be sorry. Run the engine in L5 mode. Get it to die while watching the spark plug fire. IF the ignition has quite and is causing the shut down you know where to start your real diagnosis. IF neither of these is the culprit, your going to have to start looking at whatever other fuel system components like injector power sources and such...

[dexter72]

Usually a lack of fuel causes sputtering. By not turning over at all I would look more at your ignition circuit. Ignition switch, wiring from the switch to your relays and fuses. Then check the wiring from the ems unit. If you have a voltage drop in the wiring to the starter, it can cause the no turning over issue you wrote about. A bad relay can cause the running and dying issue.

 

When the issue shows up pull a coil and have someone crank it over to make sure it does or doesn't have spark. Maybe put the timing light on it, just to make sure the timing is correct. Over advanced timing can cause the not turning over issue also.

 

Just a few things to look at.

[ktm]

Thank you to all that have responded. I was looking for confirmation of my diagnosis. It seems I am on the right track.

 

I researched this issue for 2 days and the overriding them was that a bad fuel pump can have the exact same symptoms. Runs fine, overheats and siezes. Cools down and you can fire it up again. Repeated running builds up latent heat and it runs for a shorter and shorter period.

 

When I pulled the old pump I ran a bench test to see if it would seize up or if I could hear a change in pitch, grinding, etc. While the pump did exhibit a grinding noise, it ran for a solid 10 minutes without issue.

 

After replacing the pump and seeing the exact same symptoms, I figured that it had to be on the ignition circuit somewhere. I've been sick as a dog the last two days and haven't touched the car.

 

The failure point has to be somewhere that shuts down the entire circuit. One bad coil would cause that cylinder to miss given the sequential setup.

 

The COP setup would have to have a bad common ground or bad power. There are two common grounds, one common power and 6 individual control wires. It could be the grounds, the power, the relay, or the fuse box.

 

Fueling could be an issue as well as they are controlled via a relay and all have a common power.

 

Actually, it could be any number of things along the circuits.

 

Thanks again. Looks like my multi-meter and I have a date.

Edited December 13, 2010 by ktm

[rayaapp2]

When the issue shows up pull a coil and have someone crank it over to make sure it does or doesn't have spark. ...

 

Most COP systems and some waste spark systems need to be grounded during any hot tests. KTM most likely knows this. The coils in the COP's have a much thinner/longer winding in them and are susceptible to high current when you simply unplug them and do not give them a special path to ground. Essentially you produce an over current situation when they actually do fire and the chances of you burning up or damage the coil windings is highly probable and it can leave you with a bad coil. This is why I put these instructions in bold in my first response. FYI for anyone thinking about attempting this.

Edited December 13, 2010 by rayaapp2

[mario_82_ZXT]

This sounds almost exactly like the problem my dad had with his MSnS'd l28et using the turbo dizzy.

 

I replaced the CAS (Napa sells it for a reasonable price) and while replacing it I noticed the CAS harness had some parts of the insulation broken with wires exposed. I de-pinned the plastic connector on the dizzy side and soldered new wire from that to the round connector (I couldn't figure out how to de-pin that one).

 

No problems since then (except for my dad running out of gas once).

[ktm]

Small update. I managed to work on the car for 20 minutes tonight while I felt somewhat better.

 

I do not own an O-scope, but I have put my DMM to great use. The CAS sees a full 12 volts and when rotated the fuel pump kicks on and the injectors click. I turned it slow enough not to fire the coils (they are set on a relay to trigger when the RPM is greater than 300). The voltage pulse from the CAS was around 6.8 volts.

 

I also tested the ignition switch and have continuity between the battery terminal and all of the other terminals when keyed to their respective positions. That does not rule out either the CAS or switch since they were not under load.

 

Time to start looking for grounds. If no luck, then I will circle back around to the CAS.

Edited December 14, 2010 by ktm

[bradyzq]

I would strongly suggest logging with your ECU, or maybe you've already done this. If so, what does the log say when the car dies? Or does the ECU shut off too?

[Tony D]

Bo,

One thing I notice is the thought you can overload the pump (freezes) and not blow something.

If the pump runs at full load at 10A then 25% above that a circuit breaker or fuse should stop the line overloading (and burning up the pump control relay in the process.)

 

One sure way to see if the pump is drawing down from a drag internally is to put an amp-clamp on it---if it's heating up and seizing you will see the amperage increase well beforehand of it seizing. At the point it seizes, you will see the amperage rise to locked-rotor current (3X to 6X normal Full Load Amperage.)

 

I'd look for a solid state component failing or loose wire. Good Luck!

[dexter72]

You mentioned that it would not turn over. I mentioned checking the ignition switch voltage thinking when someone says: not turn over, means won't crank. Will the engine crank over when this problem happens?.

 

I was piggybacking off of Raayaps coil spark diag. ground the plug and see if you have spark when the problem pops up. When your feeling better of course.

[ktm]

There is a great thread over at Zcar.com (they do have them on occasion) where someone had the exact same problems as I. He did own an o-scope (Tony you were advising in that thread as well) and his CAS checked out "fine". However, upon replacing the CAS his problems went away. It could be an issue with the CAS and RPM. The electronics in the CAS are 28 years old. It is located in a "harsh" environment subject to vibration and heat cycles.

 

However, a replacement CAS is $63 from NAPA and I have 30 days to return it (the shop is right next to my office). I do not want to just throw parts at it and hope something works, but it's worth a shot.

 

The car will start up occasionally. It will always engage the starter, the fuel pump kicks on and the injectors click. The fuel pressure is 43 psi and does not waver (I tested it by rotating the distributor to keep the pump running). It could still be a loose wire/connection some place and I intend to check.

[rayaapp2]

Ive had problems with the 81 style CAS... The engine would get to operating temp and the timing would go bazerk. I didnt realize what had happened until I blew cylinder #6's piston. The car was running 18lbs of boost at the time so timing was critical and it really didnt take much to pop #6. Come to find out many others have experienced heat soak issues with the early style CAS in the 81's. I have seen/used the optical setup in the 82-83 and its different than the 81's, but I do not see why it couldnt suffer similar issues from heat soak. You should be able to toss an advancing timing light(so you can do a full rpm sweep and watch the advance) on that engine setup while its running and verify your timing up until your problem presents itself again. I had to verify timing once on COPs without the funny inductive pickup for them... Use a good short peice of regular spark plug wire to gap the COP away from the plug and use the conventional inductive pickup to clamp over it like normal old school cars. That way you can verify your output without relying on the computer calculated output... This way you can verify that yeah your timing is dead nuts and it eliminates CAS, and other ignition components as the culprits and move back on to the other option, fuel. Just make sure you monitor the timing up until the point of failure to be sure that it isnt the cause of the problem.

 

You have yourself an intermittent problem so your going to have to get it to "act up" again while probing around. Hopefully Murphy wont step in and make it run perfect for the next 30 days.

 

I too believe it would be helpful if you could clarify "turn over" Is the starter not working in some way like not even attempting to click over, engage the engine, do anything, or is the engine just not firing off?

Edited December 16, 2010 by rayaapp2

[ktm]

The starter will engage the flywheel, it is just not consistently firing off. Sometimes it will fire off and then die 10 seconds later, other times it will not fire off at all. The problem is that the car will not run for more than 10 seconds so attempting to verify timing is a bit hard.

[Tony D]

Ray, that's not a heat soak issue, that's a failed CAS! That was the same thing that happened to my red 260. Get to operating temp and the CAS went nuts. It's failed. When they are new, they don't do that, seriously! But after a couple hundred thousand miles of heat cycling, just like N/A Ignitor Boxes, or turbo power transistors they fail when hot.

 

Bo, you know where timing is supposed to be: use a dial in style timing light, mark your crank pulley, and start the car with a remote starter button. As soon as it fires your timing marks will show clear as day if you use a light colored chalk on the pulley. You can see timing while cranking using this method. Turn the timing light to where you know you have it mapped, and it should line up with your indicator. If it doesn't, something has moved or you don't know your timing map as well as you thought!

[ktm]

Minor update.

 

I replaced the CAS today and the car fired right up. I crossed my fingers and was letting it run to operating temperature, but it never got there. Within 1 minute the car shut off again. I could get the car to fire up each time, have it run for 10 to 20 seconds, and then die. I pulled the cover over the ECU and watched the LED to see if there was power interruption to the ECU. The LED did not falter when the car shut off.

 

When the car dies the fuel pump gets energized again and runs for 5 seconds. I thought that maybe the fuel pump relay when bad so I replaced it with a new one I had in a drawer. The car fired up and ran for 20 seconds and died again. I managed to watch the fuel pressure gauge this time and the gauge did not move at all when the car died (no sudden loss in pressure).

 

I fired the car up again and it ran for 20 seconds. This time, however, it sputtered to a stop. I watched my AFR gauge start reading leaner and leaner. It moved to 15, then jumped to 17, 19, back to 17 and then the car died.

 

I need to take a look at my relay controlling the fuel injectors. The injectors are fired sequentially so unless all 6 control wires from the ECU or the ECU went bad, it has to be a power issue if it is related to the injectors.

 

Tony, I did manage to get the timing light on the car when it was running for 20 seconds one of the times. I locked my timing to 20 degrees in the ECU and measured the timing. It was reading 20 degrees.

 

When the car is running I can increase the RPM smoothly. There is no sputter, no missing, etc. and the AFRs are reading what they should. However, the car will just shut off whether I am idling or around 2000 rpm.