[v10projects]

Hi,

I have a 2004 V10 TDI that started acting up about a week ago.

Background: I put a NOS V10 motor with new turbos, accesories, and wiring harness in about 1000 miles ago.

After starting, it runs for 42-45 seconds (timed many times) and then throws a CEL for injectors 6-10 (5 faults, one for each injector, implausible signal, value of resistance too high). It runs rough until I clear the codes, then it starts running normally and nothing I can do makes the codes come back until I restart it. It does this probably 4 of 5 starts, whether it's fully warmed up or cold doesn't seem to matter.


The obvious suspect was the common wire from the ECU pins 31 +32 to the injectors via the connector on the back of the head, but I've checked that, all the firewall and engine bay grounds, and the ECU power supply wires. I kinda don't think it's an obvious wiring problem with how consistent the problem is, and how I can't get it to come back without shutting the engine off and restarting. I've done several hours of driving around, and plenty of idling. I've even pushed and pulled on the wiring harness to see if I can make it happen while idling, but nothing reproduces the problem except right after starting.

Maybe ECU going bad? Or something with the battery management handing off ECU power from the starting battery to the accessory battery? I think my next step is going to be trying to run the ECU with a direct connection to power and ground and see if it still happens.

I'd welcome any ideas or suggestions, I'm tearing my hair out with this one.

-Nate

 

[casioqv]

It’s possible that one or more of the injectors is failing. One interesting experiment would be to swap the left and right injectors and see if the code moves banks.

You could likely add a resistor of about the same impedance as the injectors to prevent the code from appearing.

 

[v10projects]

I can’t see how it would happen that a failing injector (or even a couple) would cause 5 separate faults, one for each injector, to happen simultaneously. Each injector is it’s own circuit. It never sets faults for fewer than all 5 injectors at once. And it seems even less likely to me that it would run perfectly after the codes are cleared. I’ve done several hours of driving after clearing the faults under varying conditions with no hint of a problem until it’s started again. I hate to open up the engine and start swapping injectors, it’s kind of a pita on these motors. Maybe alternatively I can probe a known good injector circuit on bank one with an oscilloscope and compare injectors 6-10 to it, one at a time.

I appreciate the suggestion to try faking out the ecu with a resistor. I’ll give it a try next weekend when I get back to it.

-Nate

 

[v10projects]

Well I didn't get a chance to work on it this weekend, I must have displeased the electronics gods, because my business had a CNC lathe axis drive let out the magic smoke. So the Touareg is going to have to sit until I get around to it.

 

[v10projects]

I had some time to mess with it today while waiting for parts to come in... After a couple weeks of consistent behavior, now it's doing something different.

It seemed to idle OK after starting it this morning, but when I drove it up and down the street it didn't want to maintain a steady engine speed. This seemed to get better once the engine got up to temp. While idling, engine warmed up, I got 2 faults and the engine instantly started to sound like it's running on 5 cylinders. The implausible signal codes didn't come back.

I have a flashing glow plug light, and 2 faults on ECU 2. No faults present on ECU 1.

Fault 1: 17111: RPM signal from ECU, no signal, sporadic
This fault comes back right away if cleared. Glow plug does not stop flashing when clearing code.

Fault 2: ECU processor error
I couldn't get this one to come back after clearing it.

Engine sounds like it's running on 5 cylinders even after clearing the code.


Now I'm thinking a bad ECU2 is the most likely thing, but first I'll try to check the wires between master and slave ECUs. Garbage in garbage out after all. I am now thinking that ECU2 was throwing the codes for injector signals because it had its timing confused.

 

[v10projects]

After a bunch of research, I was able to find out that the master/slave ECUs will run the engine while swapped. Each ECU will throw an implausible pin coding error, but if this is cleared from both before starting, it will run normally. I sawpped the ECUs and the problem followed the ECU. Along with the original problem. With ECU2 in the master position, I have injector implausible signal codes for cylinder 1-5. I think this is pretty conclusive.

Now I have to decide if I would rather learn how to clone the slave ecu, or just solder the eeprom into a new (used) one. I plan to do some work on my old v10 motor and swap it into something else, so I'm eventually going to have to figure out these ECUs, but I was hoping to do that when I have down time, not right now.