CCCU repair research

A

a-zwo

Member
I know the person to ask about this is probably Lau or @timmus, but does anybody have (and would be willing to share) dumps for CCCU/KSG?

As I’m sure most people are aware, some CCCU’s have flash a microcontroller which has a habit of “forgetting” its programming after many years, causing the CCCU to break.

To fix it you need a firmware dump from a good device to program a new controller with, but these dumps are fairly illusive.

Any information would be very much appreciated, but I do also understand that people make money repairing these and may be reluctant to share :)
 
Interested in the process, I was assuming the CCU issue was hardware, (mine caused a power drain, I think that's a common one). Is it actually firmware, what kit do you need to re flash it .?
 
In some cases it is hardware: Relays and CAN interfaces, have been mentioned.

But I know one of the common reasons is this MCU failure, which I suppose is also a hardware failure too.

You need to desolder the chip, use a suitable programmer which supports old Motorola chips, and then resolder it. In many cases, I think the flash in the chip is damaged, so you would have to fit a new part.

I’ve never done this, I’m just assuming the process based on the limited information available.
 
Last edited:
Pretty much, yes. But in this case it’s a MCU with built in memory, not an EEPROM.


Edit: I’ve just noticed I replied above, but it hasn’t been approved yet
 
Some more detail:

This chip has a security feature that “prevents” reading. It is a 8 byte password, that is a lot of combinations.

Most commercial programmers, as far as I can tell, use a dictionary of known keys to read the chip. I’m not sure how those keys became public domain, possibly through data leaks or possibly through defeating the security feature (there are some interesting points in the data sheet that point to this being possible in a relatively easy way, I will do some testing).

There is one mention on the German forum of someone removing a diode and that “fixed” the issue. I later found a post from Lau that clarified the diode is just part of the reset circuit for the MCU. When the chip begins to fail, the code gets into an illegal state and doesn’t trip the WDT, so that then causes the MCU to reset. No diode = No reset, but the MCU is still in an illegal state, which will have undefined functionality (the code is effectively changed).

I realise this topic is probably too low-level for a car forum, but the issue it causes does affect A2 owners a lot, and a solution could help everyone! :)
 
Last edited:
You must log in or register to reply here.
Back
Top