Intermittent 3.2 Stalling

neilschelly · 08-04-2013, 01:20 PM

I'm dealing with an intermittent stalling issue on my '88 Carrera 3.2 for the last week or so. I'm going to start tracking changes, symptoms, etc here, and either it will ring a bell for someone who can help me, or it can serve as a list of places to look if anyone in the future has a similar problem. Basically, I've got an occasional cut-out of the motor while driving. The car is my daily driver. It usually happens once or maybe twice a day, but some days, won't happen at all.

This feels exactly like when my DME relay starting flaking on me about a year and a half ago, but my replacement is fine. I resoldered both my original one and my new one to be sure, but I'm not suspicious of the relay now. It's never even warm when this happens, and it often happens on cool days without even having the car up to temp yet.

It's always been happening on low-speed backroads so far. At most, I've been going 30-40mph speeds when the engine cuts out on me. That could just be the intermittent nature of this. It's never happened more than once in a single drive, and it usually happens near the beginning of a drive. Sometimes, it's been right after another drive that has already gotten the engine warmed though, so it's not a cold-engine-only condition.

Most times, I can just restart the motor and continue on my way. About 1 in 5 times, it's taken a few tries and about 30 seconds before I can get it restarted. The starter will turn, but the engine won't catch. When it's running, it smells to me like it's running rich. There's a bit more of a burble at low throttle applications too. Voltage in the car seems consistent even when it's stalling out.

Recently replaced/serviced items:
* ~5000 miles ago: Fuel filter, Spark plugs/wires, distributor cap/rotor, valve adjustment
* ~20000 miles ago: DME Relay, but I've also resoldered my old one and behavior was the same.
* ~25000 miles ago: O2 and CHT sensors replaced, already had 2-wire CHT.

Issues fixed:
* I keep having the #2 spark plug wire fall off the plug. Doesn't seem to make much a difference, aside from the obvious issues with power and light backfiring. The new Magnecor wires are just not as nice as I was hoping for and I may just replace them outright again out of annoyance.
* I disconnected the brown wire under the driver's seat that is supposed to be just used for California emissions. I didn't notice a significant change.
* Tried putting some Seafoam into the gas tank to try cleaning out the injection system a bit. No change.
* There's a thin 6" long black hose on the back of the throttle body that connects to a vacuum line above the tee fitting that goes to the fuel pressure regulators. It was separated from that tee fitting. I suspect it has been like that since before I got the car and I've always had a bit of a rough idle. I reconnected it, and the engine started idling too low after warming up, sometimes dropping to just 200-400 RPM. With the #2 cylinder connected, it was less even, but it's usually idling smoother now.'
* I've verified current through all of my spark plug wires with a nice Agilent inductive DMM, but ultimately, I haven't been watching when it's stalled since I have to be driving then.

Theories:
* Fuel pump is failing. I cannot hear it going on and off from inside the cab. Obviously, it's usually working fine. I would expect to hear it sounding pained if it were on its way out.
* Fuel lines clogged. I'd expect this to be a be more a one-time occasion.
* Fuel pressure regulator failing. The vac line that was disconnected may have had a negative effect on the function of the fuel pressure regulator or damper, but not enough to explain this. I've read that an over-pressure condition can cause the engine to stall out and the engine electronics to cut out.
* Ignition coil failing. I haven't replaced the ignition coil, but I don't feel like this is indicative of a failing ignition coil. I imagine that would be more of a failure under higher-speed conditions when the coil is discharging at a higher rate.
* Bad fuel injectors. I've read about fuel injector coils failing and causing the fuel injector electronics to stop firing. Would this be so intermittent?
* Failing DME. I could open it up and see if there's any bad solder joints in there.

Things to investigate:
* Fuel pressure gauge - any good suggestions on a place to tap for fuel pressure? I'm wondering if a fuel pressure over- or under-pressure circumstance is present.
* Timing Light - inductive ammeter isn't a great way to monitor spark signals, but I'm not sure how much I can learn here unless I can cause the engine to stall and reproduce the problem.

So that's where I am. I don't really know my next steps because I cannot really troubleshoot much while the failure is actually happening. At best, if I run a fuel pressure gauge into my cabin, I'll at least be able to monitor that for next time there's an issue.
-N

Chuck.H · Chuck.H's Garage

classic symptoms of cracked solder joints on the DME computer - open it up and inspect the joints of the components that stand up off the board on the aluminum heat sinks.

Chuck.H
'89 TurboLookTarga, 361k miles

john walker's workshop · 08-05-2013, 12:57 PM

is the head temp sensor updated? should be two terminals in the plug. old one had one. known to cause what you describe.

Hoss911sc · Hoss911sc's Garage

FWIW I had an intermittent stalling but the cars always restarted right away - turned out to be the electronic portion of the ignition switch. The clue that I didn't see right away was the seat belt warning light coming on just as the car stalled out. The alternator light also lit up as it was stalling but adding a voltmeter in the cigarette lighter eliminated that as a cause..

Good luck, intermittent issues are always fun.

rusnak · 08-05-2013, 05:54 PM

If you have wacky electrical problems like windows not working, etc then it could be the ignition switch, yes.

If you have low fuel pressure, then you would change the fuel filter, clean the swirlpot screen at the bottom of the tank, and change or check the one way valve at the fuel pump prior to moving further.

The vacuum line disconnected on the fuel pressure regulator would make it run full rich. Your O2 sensor should compensate, but it's running at the full rich stop setting. You would then re-set baseline CO with an AFR meter.

Fuel pressure is checked at the threaded cap on the left side fuel rail next to the dizzy. A quick check: next time it won't start, try starting fluid. If it starts fine, then you may have low fuel pressure due to a clogged tank screen or check valve.

neilschelly · 08-06-2013, 06:08 AM

Quote:

Originally Posted by Chuck.H

classic symptoms of cracked solder joints on the DME computer - open it up and inspect the joints of the components that stand up off the board on the aluminum heat sinks.

I think that's an easy enough thing to check - on my list for tonight.

Quote:

Originally Posted by neilschelly

* ~25000 miles ago: O2 and CHT sensors replaced, already had 2-wire CHT.

Quote:

Originally Posted by john walker's workshop

is the head temp sensor updated? should be two terminals in the plug. old one had one. known to cause what you describe.

Yes - that's alright.

Quote:

Originally Posted by Hoss911sc

FWIW I had an intermittent stalling but the cars always restarted right away - turned out to be the electronic portion of the ignition switch. The clue that I didn't see right away was the seat belt warning light coming on just as the car stalled out. The alternator light also lit up as it was stalling but adding a voltmeter in the cigarette lighter eliminated that as a cause..

Good luck, intermittent issues are always fun.

I haven't yet been suspicious of the ignition. I know the cigarette lighter maintains power through the stall, because all gauges continue to function and the volt meter I was using to watch confirm steady voltage was plugged into it. Also, the stereo keeps playing. I'll try to pay more attention to that next time it happens.

Quote:

Originally Posted by rusnak

If you have low fuel pressure, then you would change the fuel filter, clean the swirlpot screen at the bottom of the tank, and change or check the one way valve at the fuel pump prior to moving further.

Fuel filter is new, but I'm not familiar with the swirlpot or the one-way valve. I'll have to look into those, and they sound cheap enough that it probably just makes sense to replace them.

Quote:

Originally Posted by rusnak

The vacuum line disconnected on the fuel pressure regulator would make it run full rich. Your O2 sensor should compensate, but it's running at the full rich stop setting. You would then re-set baseline CO with an AFR meter.

With the vac line re-attached, it seems like it's running richer now. However, with the #2 plug continually coming loose this week, I am more likely to believe the rich smell is just a result of the amount of fuel that cylinder is just pouring away. I really just gotta get another set of plug wires I guess - these Magnecors are just not good.

Quote:

Originally Posted by rusnak

Fuel pressure is checked at the threaded cap on the left side fuel rail next to the dizzy. A quick check: next time it won't start, try starting fluid. If it starts fine, then you may have low fuel pressure due to a clogged tank screen or check valve.

I never have so much trouble starting it that I have to get out. And I'm not too keen on leaving a can of starting fluid in my car for if it does have more issues. If I have trouble getting it to start while I'm at home, I'll give that a shot, but I think I'll just as soon replace those things anyway.

Thanks everyone for the ideas so far! Just about every time it's stalled, I've found the #2 plug disconnected again. I think while I can check these easier things, I'm just going to go ahead and re-replace the plug wires so that I can troubleshoot _one_ problem instead of two.
-N

rick-l · 08-06-2013, 06:31 AM

Quote:

Originally Posted by neilschelly

* There's a thin 6" long black hose on the back of the throttle body that connects to a vacuum line above the tee fitting that goes to the fuel pressure regulators. It was separated from that tee fitting. I suspect it has been like that since before I got the car and I've always had a bit of a rough idle. I reconnected it, and the engine started idling too low after warming up, sometimes dropping to just 200-400 RPM. With the #2 cylinder connected, it was less even, but it's usually idling smoother now.'

I'd fix all this stuff first.

When the fuel pressure regulator is working it won't idle? The idle air control valve won't compensate?

MMiller · 08-06-2013, 06:31 AM

My 88 also had intermittent running issues. Usually when cold at low rpm's mostly in first and second gear. Replaced coil, rotor, cap, plugs, speed sensors, head temp, injectors etc...

I got fed up and sent the DME to Steve Wong. Turned out to be the ignition circuit was failing in the Motronic. I think he said about 40% of them that he sees have this problem. He repaired it and so far so good....

Hope you get it sorted as I understand how frustrating it can be

Regards,

88911coupe · 08-06-2013, 06:36 AM

Sounds very similar to my problem a few years ago...turned out to be cracked solder joints as mentioned here. They can be tough to see, need REALLY good light, at least for my aging eyes.

neilschelly · 08-06-2013, 05:00 PM

I did find two cracked solder joints at one of components mounted to the heat sinks on the DME. I resoldered them both tonight and I'll reinstall it in a few minutes. Hopefully, this is enough to explain it.

I'll skip on the fuel screen and valve for now, since I can't find the valve on this site and I can't justify $50 for the screen without known cause. I will still be ordering new ignition wires soon regardless.
-N

neilschelly · 08-09-2013, 02:38 AM

2 days with no stalls is a recent record. I'm not claiming success quite yet, but I'm happy with the initial results after resoldering the DME.

I've hacked a piece of scrap metal into a hold-down for the #2 spark plug boot and attached it to a valve cover stud, so I'm pretty reliably running on 6 cylinders now too. Idle is still a bit low and the mixture a bit too rich when it gets warmed up. I'm going to want to adjust the AFM and idle to fix that I think, but that definitely seems unrelated to the impetus of this thread.

Is it really beyond the home mechanic to adjust AFM? Gotta dig into that one a bit.
-N

neilschelly · 08-12-2013, 06:21 AM

Another couple days, including lots of low-speed driving and idling at autocross events both Saturday and Sunday. I'm convinced that the DME solder joints were the fix for the stalling now. Yay!

Any my scrap metal hack has been working so well with the ignition wires that I figure I'll keep the Magnecores and just buy a set of RSR-style hold-downs for the spark plugs. I only need one, but at least I'll have more if I feel like using them.

My next projects are still to adjust idle a bit higher since it's only at around 800 when it is stable. And the exhaust still burbles a bit too much and smells rich, but again, outside the scope of this thread...

Thanks everyone for the help - I'm calling this one done.
-N

DRACO A5OG · DRACO A5OG's Garage

Oh no, if your chip is stock, then raising your base idle will do no good but make your DME over-work to go back to the spec 780-800 RPM's and may eventually damage it. Keep the RPM's where it is programmed to.

If you want higher RPMs you will need to up your chip set to the 87-89 spec or Steve Wong's 911Chips which are set at 880RPMs, he has both available.

If you removed your CAT ( for testing of course :-P ), it will always smell ( no catalyst to burn it off ) and gurgle/pop ( back pressure loss due to testing of the Cat removal )

To get an idea but not really accurate unless you use a LM1 or LM2, use a DVM to the O2S and see what reading you get after she is warm. If AFM is suspect then borrow a LM1/LM2 and correct AFR.

Been there done that...

steely · 08-12-2013, 08:02 AM

Congratulations. If you think of it, let us know exactly which of the parts it was that you re-soldered. IIRC, most of the ones that are heat-sunk are related to either the ignition, the injector drive circuits, or the fuel pump power. just curious.

neilschelly · 08-12-2013, 10:09 AM

Quote:

Originally Posted by DRACO A5OG

Oh no, if your chip is stock, then raising your base idle will do no good but make your DME over-work to go back to the spec 780-800 RPM's and may eventually damage it. Keep the RPM's where it is programmed to.

If you want higher RPMs you will need to up your chip set to the 87-89 spec or Steve Wong's 911Chips which are set at 880RPMs, he has both available.

If you removed your CAT ( for testing of course :-P ), it will always smell ( no catalyst to burn it off ) and gurgle/pop ( back pressure loss due to testing of the Cat removal )

To get an idea but not really accurate unless you use a LM1 or LM2, use a DVM to the O2S and see what reading you get after she is warm. If AFM is suspect then borrow a LM1/LM2 and correct AFR.

Been there done that...

Yes, my chip is stock. It is an '88, so it should be at 880. As I said above, I think a previous owner may have lowered the idle to correct for the disconnected vac line causing a high idle. I also think they may have just as easily screwed with the AFM, because it's just the nature of the work that I've seen evidenced on the car so far. Since reconnecting that vacuum line, it idles too low and nearly stalls out at times. I haven't removed the cats, and I'm well aware that a bit more backfire/burble would be expected in that case.
-N

neilschelly · 08-12-2013, 10:20 AM

Quote:

Originally Posted by steely

Congratulations. If you think of it, let us know exactly which of the parts it was that you re-soldered. IIRC, most of the ones that are heat-sunk are related to either the ignition, the injector drive circuits, or the fuel pump power. just curious.

Without opening it up again, I can't be sure exactly which components it was, but these on this page look like where I found two bad joints. One was definitely cracked and the other just didn't look good to me. I fixed up a few pitted or discolored ones around the board too, but not much that really was likely a problem.

DME

If I open it up for any reason again, I'll try to take better pictures and update this thread with the exact ones I fixed.
-N

steely · 08-12-2013, 10:35 AM

Thanks - does this look familiar? Not my pic.

DRACO A5OG · DRACO A5OG's Garage

Well, strange it should fight to keep the idle up to spec 880RPMs, hmmm strange.

Well, easy enough to check base idle, but only after she is nice and warm :-).

You no, "low idle" are you sure you don't have a vacuum leak, what happens when you remove the oil cap? If it does not change then you have a leak.

neilschelly · 08-13-2013, 03:38 AM

Quote:

Originally Posted by steely

Thanks - does this look familiar? Not my pic.

Yeah, I found that one online too, but since I did the soldering on the other side, I don't really recognize the bad joint from this side. I know it was one of the components on one of the heat sinks, and I think it was one of the larger transistors, but without a good picture of the whole bottom of the board, I really can't be sure.

Quote:

Originally Posted by DRACO A5OG

Well, strange it should fight to keep the idle up to spec 880RPMs, hmmm strange.

Well, easy enough to check base idle, but only after she is nice and warm :-).

You no, "low idle" are you sure you don't have a vacuum leak, what happens when you remove the oil cap? If it does not change then you have a leak.

A vacuum leak would likely give it a high idle. The oil cap "leak test" didn't point me toward anything. I've been hunting for this vac leak for 2 years, so I definitely checked for that one.
-N