Question: CAN Network Sleeping

[Citroënbender]

I’m stuck with an Alfa, which has spontaneously refused to start and run.

It has two CAN networks. The body one (doors, windows, lights, dash) is fine.

The drivetrain network is not playing. My reading of the service manual says it wakes when you turn the ignition on (which is when it bombards me with a flurry of errors).

To generalise, when a network is asleep, is there any traffic? Or is it stone-cold dead until the wake-up packets are sent?

[Franco Cozzo]

I can't help you with this one bud, it's above my security clearance unfortunately

Usually your communication goes ignition/key reader to the BCM, which then gives the OK for the ECU to let it start when you turn the key.

You might have a failed module somewhere in the CAN network preventing the network from working, that's about my only tips as this is more automotive electrical engineer territory than 'auto electrician'.

Commodores have a few common issues that cause it to immobilise itself:

1) Contact on the key wears down where it touches the key reader

2) Springs inside key reader contacts flog out and don't touch the ignition key contact

3) Faulty BCM/ECU

Cars of the early 2000s are right at the age to suffer from bad electrolytic capacitors, they leak and halal the traces on the PCB.

There was a huge drama with a bad batch of electrolytic capacitors in that era, it effects everything from car electronics to computers and the first generation Xbox.

I know an Alfa isn't a Commodore but these are all worth investigating for a no start condition/CAN network related issue.

[Citroënbender]

It’s a chip/aerial type immobiliser, the code is exchanged by a dedicated serial wire between body module and PCM. Apparently this exchange is not being completed; it’s hard to fathom where the dialogue stops.

Do alternators ever give a “kick” on key shutoff? This car is on its second set of modules after the first set lost the plot, and I’m now wondering if it’s spiking the system. It’s an eight year old Bosch (Europe) alternator with about 35-40K on it.

[Franco Cozzo]

You can get high voltages induced when magnetic fields collapse through a coil, that's why relays usually have diodes or resistors across their coils to prevent voltage spikes.

You won't get it on an alternator because it has rectification diodes and also a big *** battery in the way.

May be worth getting the alternator on an alternator test bench, but you'll be pushing **** up hill to find a workshop with one of those in this day and age I reckon.

I had one as old as GasOlane and used it like twice because they were never worth rebuilding.

[Citroënbender]

Cheers, I thought the battery might damp any spikes. It takes about three weeks to remove the alternator so I’ll put that into the “try to avoid” box for now. Actually know where an alternator test bench lives, and use it on my rebuilds. Really hoping not rats/mice in the firewall insulation, as a cause.

[NeillNZ]

Originally Posted by Citroënbender

I’m stuck with an Alfa, which has spontaneously refused to start and run.

It has two CAN networks. The body one (doors, windows, lights, dash) is fine.

The drivetrain network is not playing. My reading of the service manual says it wakes when you turn the ignition on (which is when it bombards me with a flurry of errors).

To generalise, when a network is asleep, is there any traffic? Or is it stone-cold dead until the wake-up packets are sent?

So is there just no activity at all on the drivetrain CAN? Apart from the obvious fuses and dead ECU it could be a node on the network has a dead transceiver. A dead transceiver can drag the network down effectively creating a short circuit. The way to find this would be isolating the various nodes one at a time. What about isolating the ECU and seeing if it is making signals?

What tools do you have? An oscilloscope? CAN to USB dongle with software? Perhaps PM me..

[Citroënbender]

G’day Neill,

It was a failed major cap in the ECU, basically every time the unit woke and tried to initialise the throttle plate, it exceeded available power. This gave me a rapid cycling of throttle flap motor and fuel pump.

As the fault was a known issue for this family of ECU, I tried a moderate “percussion test” with screwdriver handle, rapping on the case. That did the trick - it immediately went cold stony dead. Fixed now, just over $500 later. Although I knew the fault and its solution, lacking experience, a “clean room”, specialised gear including a scope - it made good sense to outsource.

[Franco Cozzo]

Originally Posted by Franco Cozzo

3) Faulty BCM/ECU

Cars of the early 2000s are right at the age to suffer from bad electrolytic capacitors, they leak and halal the traces on the PCB.

There was a huge drama with a bad batch of electrolytic capacitors in that era, it effects everything from car electronics to computers and the first generation Xbox.

I know an Alfa isn't a Commodore but these are all worth investigating for a no start condition/CAN network related issue.

*Flexing bicep emoji*

[Citroënbender]

It was simpler than that, amico, the cap just falls off as its solder glue fails from heat/vibration.

Link to explain better: https://www.alfaowner.com/threads/14...repair.375231/