Ghia

Hi all,

I haven't been on here for a while, busy with projects

I currently have a car in which has an intermittent fault.

It's a 1989 944S2. Driving along the road normally in traffic, the engine suddenly loses all power and dies. It will barely tickover but will rev up to about 2,000 (not misfire) if you hold the throttle just off idle.

This fault has been plaguing me for a while now and I've done the following:

Replaced the following parts:

ECU
Air Flow Meter (MAF) twice, one 2nd hand, one reconditioned.
Hall sensor
Coolant Temp sensor
Fuel pressure regulator
Crank timing sensor
Coil
HT cap and leads
Plugs
Rotor arm
Throttle Position Sensor

It does not have a Cat or Lambda Sensor and has the jumper plug in where the lambda sensor would plug in.

I've checked the fuel pressure while it's faulting, No problems.

Earth connections are all clean and tight.

I did a pressure test of the intake system thinking that I might have an inlet leak somewhere. Everything held up fairly well, to the point that oil starting weeping out of some of the small oil seals on the rocker cover. No obvious pressure leaks.

I've thought about the petrol tank breather valve... but it was removed before I got to the car and the pipe has been blanked off on the inlet manifold.

The most recent addition is that I discovered that the coding plug next to the ECU was missing. I replaced this and this has now changed the intermittent nature of the fault so that it fails after roughly 5-8 minutes of running.. i.e. it's warmed up and is looking for a reading from the lambda sensor (just my theory here).

Something else that you should know. If the fault is occurring and I disconnect the lambda sensor bypass plug, the fault stops for a short period before faulting again. The same happens when I plug the bypass back in.

If I remove the coding plug by the ECU, the intermittent nature of the fault returns.

I plugged a Snap-on injector tester into one of the injector plugs and tested it again (obviously this creates a misfire as one of the injectors is disconnected). What I found is that the injectors are losing current when the fault occurs. Not a total lack of current but just as if you were decelerating.

I've read the workshop manual many times and tested and retested as many items on the engine as I can think to.

I don't have access to the Porsche diagnostic tools and I can't find a diagnostic plug in the loom anywhere so can't retrieve anything from the ECU.

I'm getting ready to change a wiring loom.

Any ideas?

MG

Ghia

Oh, I forgot to add, the coding plug is for RoW to run without Cat.

Coding plug part no: 928.607.422.00

MG

johntorg

I believe that, in the DME, there are two driver's, each one firing 2 injectors. You might be running on 2 cylinders if one of the DME wires is intermittently open. Check each of the injectors to see if all of them lose power when the fault occurs. If 2 of them don't, that is probably your problem. Lindsey sells a kit to replace just the injector harness.

Ghia

Hi John,

Thanks for the feedback.

I'll look into the loom. When it faults, it still runs smoothly on 4 cylinders, just won't rev.

If it were a carb'd engine, I would say that it was running out of fuel or that the main jet was blocked.

MG

tod84944

Check fuel pressure when you have the bad running. I had a customer with a S2 with the exact issue. Traced it to a soon to die fuel pump.

Ghia

I've changed the fuel pressure regulator and had a pressure gauge on the engine while it faults. The pressure doesn't change and is spot on.

MG

Makis

I would check the throttle mechanism. The position sensor I think is not enough to change also needs adjusting. It should click when you operate the throttle by hand from the idle position. Also get all the connectors you see in the engine bay clean if you can. Disconnect them, WD40, reconnect. AFM connector is prime one and is easy to remove/connect.

Ghia

I checked the function of the TPS from the ECU plug to check that the continuity was correct at idle, partial and above 2/3 throttle. This way, I checked the loom as well.

It is possible that the loom to this may be faulty. I just don't understand why it would fault so regularly after 5 minutes or so of running.

MG

tod84944

In your list of parts you replaced, you have temp sensor. Did you replace the DME temp sensor or the gauge sensor.

Ghia

I changed the engine temp sensor as can be seen in the centre of this photo with the blue plug attached.



I also checked the ohms reading across the ECU pins as per section 24/28 in the manual. I did only check this once while the engine was around 30oC.

MG

johntorg

How did you check that the engine was firing on all 4 cylinders when the problem occurs?

Ghia

I haven't done any testing to check that, just my experience of knowing by sound and feel that an engine is or isn't mis-firing. I come from the days tuning carb engines by ear.

Over the last few months, I've been working through this fault on the car and gradually replacing parts. Because of the intermittent nature that the fault used to have, I've spent some time just driving the car to and from work just trying to get it to fault so that I could try to diagnose the fault.

When the car first faulted, I found that the engine would continue to run on full throttle or at idle but had nothing inbetween. This is the reason that I went with the TPS as one of my first things to change. It never mis-fires, is always smooth. The power just dies completely like you've turned the engine off.

As I changed the parts named above, I found that the engine will no longer run on full throttle when it faults. Only idle and just above.

MG

gtroth

Boy, sure sounds like AFM to me, but not for 3 different units (was AFM for me with similar severe hesitation when car was warm and air was very cold). I'm also a little suspicious that it's not set up properly for no cat operation since I don't really understand why it does what it does when you unplug an rep lug things.
Couple of other things that came up in my debug thread were ICM, coil, and ref/speed sensor. And flaky AFM connector (clean with contact cleaner).
These are tough. Intermittent and you don't even know if it's fuel, spark, or air. Good luck - I'll chime in if I come up with something.

Ghia

Thanks for the link and comments. I'll have a read.

I have also changed the Ignition Control Module - there have been so many parts changed that I'm losing track of them... But also running out of things to change!

I didn't like the look of the wiring going to the AFM and changed that fairly early on, all connections were soldered and heat shrunk.

MG

peanut

on the face of it this sounds like a problem that should be now easy to track down . lets face it you have replaced practically everything now, bar the loom!
you've eliminated a lot of stuff so you should be getting closer to solving this


You have replaced the ECU without improvement so it should be ok to dismiss that as the direct source of the problem however everything to do with the fuel delivery and ignition pulsing is controlled by the ECU which in turn is controlled to a large extent by the various sensors so it could still be the indirect culpret.


The ECU not only earths the injectors and ignition amplifier but also earths one of the DME relays internally so if for example the earths of the ECU were poor or intermittant the ECU would periodically fail to earth the DME relay so the ECU would lose not only its own power supply but also the fuel pump, injectors and ignition amplifier supplies would fail also

The symptoms would be that the fuel delivery was faulty ie faulty pump,FPR, blocked filter, sticky injectors etc etc whereas the true fault was electrical ie bad ECU earth or DME relay earths

To ensure that the DME is not the source of the problem i would advise that you do all testing and running with a 3x wire bypass fitted. This will force the operation of the fuel pump and provide the ECU ,injectors and ignition with a constant 12v+ supply at all times. At least you should not have an intermittant DME relay operation to worry about. If you fit a simple switch in the wire between terminal 30 (constant live) to the other 2x wires (terminal 87 & 87b) you'll be able to switch the bypass on and off without having to take the damn thing out every time.

Next I would check and clean the 2x critical earth points MPll & MPlll which ground the ECU and DME relay . They are located under the dash and next to the fuse/relay box. (big brown wires)

It might be worth substituting a good ignition amplifier to eliminate that. Its bolted to the wing panel in the pass headlight well.As you say it looks like the problem is heat related and the IA runs hot ! which is why it needs a good heatsink fixing to the wing

Have you checked the hall sensor behind the distributor ? ....wiring, electrical connector etc ?

A couple of other obscure intermittant problems that occur are blocked fuel tank vent and intermittant vacuum ie brake master cylinder diaphram .

I think the fact that the engine will both idle and rev at full load but not in between is very diagnostic but I'm not fully awake yet and need a coffee so I'll come back to this later