Short on Hall wiring causing ECU ICM Memory Error?

[irishs2]

Ok doing a bit of daily driving today and my spidey senses started to tingle. I thought I noticed a miss, but was hard to know for sure at motorway speed.

Not 3 mins later, the engine died while on a roundabout. Luckly while still rolling I was able to start it again and was trouble free for another 2hrs of driving which including some

At home tonight with vagcom I get the following:



Cleared the codes and all is well. Engine starts but was a little bit slow to start with VAGCOM connected, starts faster with it disconnected. Maybe thats my immagination.

Would I be right in thinking that maybe I have a short to ground on the hall wiring and its causing the memory error, or is this the usual error I would get if the hall was failing?

[prj]

Internal memory error = a byte was misread from one of the chips.
Usually a sign that your ECU is about to go pop, but you might have caused a heavy voltage spike, and it might be fine.
Drive it around then pull codes again.

[irishs2]

Thanks, really didnt want it to be an ECU issue.

Luckly this was a repeatable error. For several weeks I have been able to get the cleared error back after 5-10miles of driving.
I have changed the camshaft sensor , and the error has not returned so far so good with 150miles done.

[UrS4boy]

Just to be clear for future reference, if you pull the codes when the engine is stopped you will always get a G28 engine speed sensor error (e.g. the 00513 code). That is a false positive. The sensor is likely fine, the engine is just stopped, hence no signal.

I trust replacing the G40 cam position sensor is the answer to your problem.

[irishs2]

Thanks, Yea G40. I was worried about that memory error , but it has not showed up since changing that sensor.
The car cutting out in traffic was not good at all , glad it's behaving now.

[UrS4boy]

Ok doing a bit of daily driving today and my spidey senses started to tingle. I thought I noticed a miss, but was hard to know for sure at motorway speed. Not 3 mins later, the engine died while on a roundabout. Luckly while still rolling I was able to start it again and was trouble free for another 2hrs of driving which including some driving.

I have heard of your symptoms, i.e. engine dieing while driving, before and the solution turned out to be the G40 Cam Sensor. In theory, this should not be the case because in theory, once the engine is started, the ECU will use the G4 crank position sensor to do the math if the G40 fails. However, when you go to restart, without the completely failed G40, the ECU will never restart the engine. It seems like this intermittent short to ground defies the conventional (and VAG) "logic".

Good use of VAGcom to track down the problem and lead to the solution, i.e. a new G40 cam sensor.

[irishs2]

Yea i was thinking it's only certain types of g40 failure would cause that symptom, and as you say that short to gnd causes the engine to stop. This behaviour is very like a faulty ignition switch . Which I had ruled out with a spare switch .

[UrS4boy]

Yea i was thinking it's only certain types of g40 failure would cause that symptom, and as you say that short to gnd causes the engine to stop. This behaviour is very like a faulty ignition switch . Which I had ruled out with a spare switch .

I hope you don't mind but I have shared an edited version of your story on my home forum because it is the first documented proof of that mode of G40 failure that I have seen (the VAGcom screen shot was the clincher).

http://forums.quattroworld.com/s4s6/msgs/139944.phtml

[irishs2]

Absolutely work away.

I can add a little bit

I put a scope on it before I changed it and this is a rough drawing of what it looked like



So I think the jaggies are moving up to ground when it should be a -ve signal, and so occasionally it did go to ground and trigger that fault. As best as I can understand. I hope that makes sense.

[UrS4boy]

Absolutely work away. I can add a little bit: I put a scope on it before I changed it and this is a rough drawing of what it looked like. So I think the jaggies are moving up to ground when it should be a -ve signal, and so occasionally it did go to ground and trigger that fault. As best as I can understand. I hope that makes sense.

Fantastic addition. Shows again what that intermittent G40 fault does. I trust Paul N sees this thread.

And just to confirm, I understand that after you changed the G40 and cleared the codes ALL of the codes, including the ECU error code, went away (??).

[irishs2]

Yep all went away except the usual G28, and I have been actively chasing the error, checking everytime I come back from a trip, had a long trip yesterday after confidence was built up again and still no sign of the error.