ACEparts_com

I agree with the logic as there's Soooo many things that can cause the same symptoms. Everything would be very simple to diagnose if things completely FAILED. As it is, many items break down, we have 15 year old connections to deal with, alarms and other wiring modifications thrown in to the mix too. It's a wonder our cars ever run.

https://rennlist.com/forums/964-foru...lywheel+sensor

Same symptoms, smell of fuel (now I don't know how much fuel needs to be unignited to get the smell, not a lot i'm guessing)

Droops83

The fuel pump will run for a few seconds after the engine is first cranked to prime the system, but will stop if there is no RPM signal from the DME, to prevent flooding, etc. So, if you are repeatedly cranking it trying to get it to start, like that 964 guy was, I could see there being a raw fuel smell from the small amount of fuel injected on the first few seconds of cranking.

In your case, it sounds like raw fuel is continuously being dumped in there while cranking, and it seems the same thing happens when it starts to sputter and die, from running excessively rich. I don't see any "normal" sensor failure causing this; something is shorted in the wiring harness, or a DME issue, or something. Sorry that I cannot point to anything more specific but I know that a engine speed/reference marker sensor failure will NOT cause your symptoms. You guy will have to do more troubleshooting. A Hammer or PST-2 scan tool would be very helpful.

Good luck.

budge96

" Or Something huh?" Well poo poo motronics if you want to but the DME will not signal the pump to start unless its getting
some signal from the reference sensor.
The ignition in turn receives its signal as well from the speed sensor to time the spark,if either of these
values is skewed or off or non existent in any way you get a no start ,DME or ECU fail symptom.

Droops83

The speed and reference sensor are one in the same in the 993. There are 2 missing "teeth" on the flywheel that act as the reference mark that TDC on cylinder #1 is coming up. The speed/reference mark sensor uses the other 58 teeth to determine the engine RPM.

So, if that sensor is bad, you get no speed OR reference mark signal to the DME, and the fuel pump will not run for more than a couple seconds while cranking. So yes, a bad sensor would cause a crank but no start condition, but in ACE's case, he is CONTINUOUSLY getting a lot of fuel injected while cranking, and even when it is running, it will randomly start running pig rich and sputter, and eventually die. A bad speed/reference mark sensor would not cause this, that is my point, I don't want him replacing parts unnecessarily as he already did by replacing the fuel pump.

Like Kevin said, I would want to be looking at live data on a scan tool at this point to know what is going to when the fault happens. Anything else is shooting in the dark.

ACEparts_com

Well, with all intermittent faults; only time will tell. Unfortunately there's a whole host of components that could cause these symptoms.
I'll post my findings as I always do in order to help others in the future.

Felix

Alan, I have a PST if you get stuck.

Taj

Alan, I can't remember but have you tried swapping out the ECU with another one? An independant should have a spare... Northway do.

cheers

ACEparts_com

BTW, using the fault code reader (can't remember it's name, the one everyone uses connected to the PC), DME voltage was the only thing recorded once from memory.

Can anyone tell if the 993TT has a camshaft position sensor or distributor sensor as the flywheel sensor cannot be the only thing telling the DME about rotation as the flywheel cannot determine if it's a power or exhaust stroke.

Felix

There's a sensor in the dizzy.

ACEparts_com

So it's plausible (mythbusters) that with a 'dodgy' flywheel sensor, the DME could still be getting enough info to fire the fuel injectors from the distributor and flywheel pulses without receiving the TDC signal from the flywheel to actually fire the plugs, or at least at the correct time.

Like I've said, time will tell.

I know that when my AC died it took me two days of diagnostics to figure it out! The wiring diagrams are not actually accurate in as much that the heater CCU sends a signal to the DME that then sends a signal to the AC relay. The diagram shows that the DME is only receiving a signal, not giving it out, but that's another story....

Felix

I would have thought that scenario would certainly generate an error code if the engine would run at all.

Droops83

As I said before, I highly doubt that the flywheel sensor is causing this. Even if you do suspect it is bad, It can be scope checked if you have access to a DSO. Even easier is next time it cranks but does not start (which is the frustrating part!) you can check to see if there is spark; if there is none you can follow the factory test plan to find that cause (or course the intercooler being in the way makes such tests more of a pain!)

It sounds like your problem is intermittent over-fueling, which I don't think any "normal" sensor failure would cause: even if the cylinder head temp sensor were bad (open) and and made the engine run full rich, the "normal" range of full rich would not cause the engine to sputter and die; it may hunt a bit if the CHT were bad but I don't think it would kill the engine or cause a crank/no start and flooding . . . .

Same with air mass sensor or oxygen sensors (O2 sensors are ignored on cold startup), which are the other 2 major engine sensors that help the DME calculate injector pulsewidth, I just don't see the usual failure modes of any of these sensors as the cause of this (unless your fuel system is modified with large injectors or high flow fuel pump???)

The point I'm getting at is, while these pesky intermittent issues can be frustrating, you can at least attempt to eliminate the very unlikely causes of the issue and narrow it down to what it could be. If you can get your hands on a PST2, you can watch the live data in graph form and with any luck you can see which numbers change and how if the engine happens to start stumbling again (I'd be looking at the injector pulsewidth, O2 sensor signals, fuel trim, airflow numbers) and see what changes, and that should help narrow down the cause. I'd put a fuel pressure gauge on there for fun as well, it's a long shot but a stuck closed fuel pressure regulator or kinked return line can cause very high fuel pressures and rich running, but I doubt that would cause your particular issues . . . i'm still betting on DME and/or harness issues.

Anyway good luck and report what you find.