3.2 Engine Stalling - Stumped

[dfink]

I will once again offer to send you 6 injectors that were removed from my 1984 engine when I upgraded to ITBs. I can also include a working motronics unit and DME relay along with a fully functional distributor. In essence the entire control system minus the coil. If you really want to go for it I will send along very nice intake control box. If you can't make it run with that then clean the dang grounds. If you find a part you can use either buy mine, send mine back and get a new one or get one from someone else. I don't care as long as you send my stuff back once the car is running. No need to spend a bunch of money just to test a part.

[TibetanT]

Scott:

First, sorry to hear of your problem with your car stalling like this.

Second, I am no expert but this sure does sound like a problem I had with my 1987 944n/a since they share some of the same Motronic components. In my case, when I installed the new TPS (Throttle Position Switch) I did not set it up correctly for FOT (Full Open Throttle) so it did exactly what you're describing here stalling/cutting out at 3000RPM. However, with that said it appears that has been looked into.

The other thing that comes to mind, when these types of "hair-pulling" problems that just don't seem to get resolved happen, is: Is it possible this car was exposed to water or salt water that got into the wiring in the tunnel? It seems to me that these types of problems appear after the wiring begins to corrode.

Anyway, hope this helps.

[Blackie911]

Originally Posted by EMJ

+1. I had the exact symptoms and it turned out to be cracked solder joints on the DME. Took all of 10 minutes to solder them up and 5 years later, I've never had this problem again.

+2

This procedure is worth doing just on general principles. You can also have a Very competent electronics shop with STRONG magnifier reflow solder joints especially around the transistor. When you button it up- mount the DME on a thin sponge to absorb some of the vibration.

[wnsgc]

Greetings All,

Dfink, I do appreciate your generous offer to let me try your injectors and DME. Since I had my DME fixed by Steve Wong, who placed it in one of his 911's afterwards, I believe it to be good. Also, I did try two other known good DME's and countless relays, all of which have tested out to be good. As for the injectors themselves, after last nights tests using an oscilloscope, we're pretty confident the injectors are indeed the problem, thus I'm going to order a set today from MotorMan (a guy who many folks on this forum have recommended). I'm not sure if you saw this on earlier threads, but my mechanic has a stock-pile of new-old, new and known working used parts, thus he's tried many of the parts I've mentioned (temporarily) just to see if the problem went away. In reality, the only parts I've replaced on a permanent basis are the sensors, fuel pump, and the harnesses (because we smoked my wiring harness during tests). The WOT switch and throttle body were replaced when I installed a Steve Wong chip, about 18 months before this issue started, thus to TibetanT's comment, which I appreciate, that is probably not the issue. Over the past few days, we've done some pretty good testing on this area of the car/engine, and if you think about it, from the DME to the Injectors, there's not much in between. Since we know the engine harness is good, and we know the fuel injector harness is good, and we know the DME is good, what's left?

Blackie911,
I like the idea of the thin sponge and will try that.

TibetanT,
There's no evidence the car was exposed to water at any point, as there's no rust or corrosion any where on the car. Also, this is the third wiring harness in the car, so it's tough to believe they'd all have issues, but we did check this one out thoroughly.

Also, since the car is 27 years old, it may be ready for new fuel injectors. Of course, I've been saying that about everything else we've tried to date... :O

Thanks again for the input and suggestions, I'll let you know what happens after I install the new injectors.

[rick-l]

Damn... No scope photos....

[bigel]

I have a similar issue with my 87 targa; the former owner is a good friend and confirms the issue has been with the car for the 16 yrs that he owned the car.

The issue is a hesitation that occurs approximately 5 minutes into driving, it lasts for ~30-60 seconds (seems like 15 minutes when you're traveling on a highway), primarily when the car has sat idle for an extended period. The car really bogs down and seems to be starved for fuel; the odd thing is it will not stall, but it seems as though it will at any moment.

The issue has been there with a stock DME and was not remedied with the addition of a performance chip, so it's obviously not a DME issue.

Puzzling...

[rick-l]

Originally Posted by bigel

The issue is a hesitation that occurs approximately 5 minutes into driving, it lasts for ~30-60 seconds (seems like 15 minutes when you're traveling on a highway), primarily when the car has sat idle for an extended period. The car really bogs down and seems to be starved for fuel; the odd thing is it will not stall, but it seems as though it will at any moment.

The issue has been there with a stock DME and was not remedied with the addition of a performance chip, so it's obviously not a DME issue.

Puzzling...

Holy buy some tools batman... That could literally be anything including the DME.

What does the tach do while this is happening?

MacGyver would rig a fuel pressure gauge he could watch inside the car.

Injectors here would be a $200 gamble.

[ischmitz]

Sorry - was busy with stuff. Here are the promissed scope shots. Negative lead to signal (pin 1 for ignition, pin 14 or 15 for injector) and positive lead to B+ (pin 18) on the DME.





Ignition coil: DME turns on (red). The coil "sees" full +12V for about 4 milliseconds of dwell time. Then (green) the voltage across the coil is reduced to 5 volts reducing the current while holding the magnetic field. At (blue) the coil is turned off and the collapsing magnetic field induces a voltage spark that gets transformed to release a spark.






Two shots with different time base of the injector. Yoy can see the peak&hold pattern: At (red) the DME turns the injector on with full +12V for 200 usec. in "peak " mode. This is to shock the injector open. Then after a short duration the injector is kept open in "hold" mode for a total duration of 1.7 milliseconds. The "hold" mode beginning at (blue) and ending at (orange) is a modulated voltage. The injector sees an average current that is equivalent to the duty cycle. This is called PWM (pulse width modulation) and is a common way of controlling currents throgh inertal electrical devices. The current is monitored and the duty cycle adjusted accordingly. At the end of the injection the magnetic field collapses and induces a spike into the negative that has to be handled by the DME.

Let me know if these help and what comes out of testing.

Ingo

[Otto]

"Injectors here would be a $200 gamble."

That's exactly it! The likelihood that all 6 injectors have coil shorting problems is very low.
And if that were the case, only buying new ones would solve the problem, as any "rebuilding"
(basically a cleaning) would not resolve an intermittent coil shorting. Again, this thread leads
to another shotgun troubleshooting effort as is usual!

If it's thought that some injectors may be shorting, then selectively remove pairs of injectors
and then duplicate the failure mode. Also, use a noid in the place one injector to easily
determine if the injector drive signal is still occurring.

Again, monitor the fuel pressure to eliminate a possible bad fuel pressure regulator!

[rick-l]

Thanks for the scope photos. I've never had the need to look at the fuel injector waveforms so hadn't really verified what they were. Now if only you had a DC current probe.... just kidding.

You should point out that if he doesn't have a fancy scope meter and and looks at the signal referenced to chassis he will see the inverse of those (or at least 12-those)

Why are OEM injectors for this car $426 each? If you are going with the generic (different part #) Bosch injectors I would definitely do all 6. With all these intermittent stories that are springing up I am nervous about my reconditioned ones.

[wnsgc]

Greetings,

I ordered injectors from MotorMan today, they are Bosch reconditioned injectors, and came in a bit over $200 for the 6. So yes, I'm taking a gamble, which I hope pays off. They come with a one year warranty, so only time will tell. As for Otto's comment, we have checked the fuel pressure and it was between 35 - 40 lbs, which I'm told is correct.

Last night, we hooked up an O-scope to the injectors and when the car died, the injectors flat-lined. Also, removing one injector allowed the car to restart, but upon reconnection, the engine died. If we consider the flow from the DME thru the wiring harness thru the engine connector thru the fuel injector harness to the injectors, the only thing that hasn't been replace and/or tested is the injectors.

They should arrive next Tuesday, so hopefully a day or two later, I'll have more info to report. By the way, I have a video of the o-scope. Can someone tell me how to place it on the forum?

Thanks.

[rick-l]

Originally Posted by wnsgc

By the way, I have a video of the o-scope. Can someone tell me how to place it on the forum?

utube account?

[Otto]

"we have checked the fuel pressure and it was between 35 - 40 lbs, which I'm told is correct."

Based on the unique failure mode, the fuel pressure test needs to be done as follows:

1. Connect the gauge PRIOR to the failure.
2. Leave the gauge connected.
3. At the time of the failure, immediately check the fuel pressure.

Any test of the fuel pressure without the gauge always connected and NOT disconnected
prior to the failure is meaningless to test for an intermittent high fuel pressure!

[wnsgc]

Otto,

In your opinion, what would cause the fuel pressure failure?

Thanks.

[rick-l]

Otto

There are two threads here.

wnsgc and RPSTech are troubleshooting an 85 and bigel kind of hyjacked with an 87.

[ischmitz]

Originally Posted by rick-l

Now if only you had a DC current probe....

How about a 0.1 Ohm resistor in series with the coil or injector. And with the DME open I just measure across the shunt resistor. A simple Fluke DC current probe is 600$ - yikes...

[techman1]

What I am remembering about the injectors is they get a ground signal from the DME to fire? Or was that something else?

If so, I would wonder if the DME harness is not giving a good ground to the DME. On Steve's cars the harness is good and gives the DME a good ground to send, and that portion does not fail. If I am wrong, back to square one.

See post 25 in this thread. Motronic Injector Signal Testing

[rick-l]

Originally Posted by ischmitz

How about a 0.1 Ohm resistor in series with the coil or injector. And with the DME open I just measure across the shunt resistor. A simple Fluke DC current probe is 600$ - yikes...

That would be a lot of work just to find what the peak and hold current levels are. You would be able to see how long they took to open by the current discontinuity.

I think I calculated the current set point from that roll your own op amp (diff pair) for the ignition coil at 8 amps.

[bigel]

Originally Posted by rick-l

Otto

There are two threads here.

wnsgc and RPSTech are troubleshooting an 85 and bigel kind of hijacked with an 87.

No Rick, there's only one thread...Bigel was simply "kind of" interjecting his similar experiences, though he happens to drive an 87.

[ischmitz]

Rick, not a lot of work at all. Here is what I do:

I insert a 0.1 Ohm resistor into the line between the injectors and the DME. I can measure the voltage drop across that 0.1 Ohm resistor. The current I through the resistor (and the injectors) causes a voltage drop U. Ohm's law gives me I=U/R. So I multiply the measured voltage drop by 10 (divide by 0.1) and that is the injector current. BTW, the shunt in the 3.2 DME for the injector happens to be 0.1 Ohm exactly.....