2007 Norge - ECU 44 code

I would change both the start relay and the start assist relay

Relays are not expensive -

Bit of an update on this

Decided to start from the beginning and re-check everything from battery to ECU.

Pin 87 on the Injection relay - 12V,  relay was also function-tested off the bike
Pin 2 on the Starter relay - 12V, relay was also function-tested off the bike
Pin 1 on the ECU blue plug (left side as if you are sitting on it) - 12V wired from Pin 2 of the starter relay.

All wires tested for resistance with nothing unacceptable. ECU ground wire thoroughly cleaned, ECU body cleaned where it touches the ground wire. ECU shows strong continuity to ground with low resistance (used the threads on the tank bolt-hole to check.

The only thing left that it could see it being now from this wiring diagram is the ECU itself.

I am going to call an MG ECU specialist on Monday and explain the situation / my testing. With some hope they'll be able to provide at least some verbal confirmation that this could be an ECU fault, to which I'll send it to them for testing.

Taking another stab at this although it does sound like an ECU issue.
I found what might be the correct schematic for your Norge.
Based on your posts, you've checked the coil circuit for the start relay. have you checked the starter side of the circuit?
pin 3 of the start relay should have 12v coming from fuse B(15amps). and pin 5 should go directly to the starter solenoid

Fair question though unfortunately it's not yet a factor

Pin 3 is actually the wire you alter during the fix for "startus interruptus" so that's currently its own fused wire taken direct to the positive terminal. However, did check for clarity and read strong 12V.

I think I understand in concept what's happening, which is why it's led me to the ECU. Key on triggers the main ignition relay which powers one half of the starter relay coil. The other half of the coil goes into the ECU, which waits for the starter button signal on pin 27. When pressed, the starter relay should be engaged by the ECU and the bike attempt to start. However, the relay doesn't engage (no audible or felt "click" despite verification of the relay's function and the use of another relay). ECU is reading the voltage off pin 1, and somewhere between pin 1 and ground the ECU is losing too many volts and refusing to complete the relay circuit. Other norge owners traced their issue to said ground however, I feel confident I have ruled that out.

Speaking my dad a little more, we both recalled that this happened last year too during winter. However on that occasion I fixed it with the startus interruptus wire. My hypothesis is that either the temperature, moisture, or both of UK winters has compromised the ECU in some manner.

We'll see how my call on Monday goes.

RESOLVED

But in I must admit, a very unsatisfying manner.
Sent the ECU off to be looked at, to a company that stated on their website they do the specific model on the norge.

Test comes back inconclusive - Turns out they just assumed they could test it because it's the same model used in some Fiats...

Gets sent back but can see it was opened and looked at/resealed. At this point I'm thinking if it's not the ECU it must be a wiring short on pin2 of the relay (after pulling the ECU code into guzzidiag I was getting p0170 which suggests a short in that circuit)

Reinstall the ECU and give it a hail mary before diving in with wirecutters. Starts first time with no issue. Repeat a few more times for confirmation.


Clearly something wasn't right in the ECU, and it being opened up jostled whatever it was back into position. Hopefully it holds up but I'll know where to look if this crops up again!

Love a happy ending but unsatisfying as you say. *Something loose apparently or not seated well.

I know of nobody who has ever managed to disassemble a W5AM ECU. Those that have tried have invariably found separating the bonding agent destroys the unit. 

Glad it's fixed.

I think all they did was take the top cover off held by the two pins, judging by where I can see the re-seal job.

Getting it back together this weekend after another test to get her running again

Fingers crossed that it's a long term fix!

Pete Roper wrote:

I know of nobody who has ever managed to disassemble a W5AM ECU. Those that have tried have invariably found separating the bonding agent destroys the unit. 

Yes.....



[You must be registered and logged in to see this image.]

Yeah that's a bit too much



Going off this video, and where I can see it was touched I'd say they got to about 2:27 in the disassembly process and took it no further.