Wireless Di2 -- Need to vet ideas

[andy2]

We have hacked it and it is NOT a canbus system. The 'consencus' is just a bunch of internet speculation. (It is a frequency (MHz region) modulated pulse code carried over the +-7.4 volt if you want to know). It simply cannot be canbus since canbus by definition needs more than 2 pins....
Typically:
pin 2: CAN-Low (CAN-)
pin 3: GND (Ground)
pin 7: CAN-High (CAN+)
pin 9: CAN V+ (Power)
If it had been canbus, or any other such standard or derivative thereof, it would have been fairly easy to crack using off the shelf analyzers and software. In the automotive industry this is pretty much standard practice. The Shimano system is entirely proprietary and non-standard and much better suited to bicycle application; a canbus system would highly impractical (or useless) in a bicycle since it has to be terminated at each end with 120Ohm resistors and thus draw a lot of power.

[kulivontot]

Thanks for sharing your knowledge, I've not touched a di2 system, I'm going entirely off of what others have discussed. If you've analyzed the protocol, can you provide some oscilloscope screenshots to help others decode the Shimano protocol?

[andy2]

Errr... No! Shimano would take a dim view of that I'm afraid. We looked into how it worked simply because we felt we could, our knew people that would find out very quickly to be correct;).
We just like to find out how things work. Finding out the workings took less than 30 minutes, the really hard slog is to catalogue the code and learn what impulses do what - to learn the e-tube language. You then need to design and manufacture the electronics for the host processors that broadcast the messages across the system.