• Di2 project for the TT bike.

    I was doing some reading about di2 hacks, seen the guy that managed to partially reverse engineer the canbus protocol of it, the guy that printed custom circuit boards for it...etc. The thing is, on a TT bike its a huge convenience if you can shift both from the base bar and the extensions, so why not try to make that work? The biggest prohibitive cost for me was the cost of cables. You can always just wait it out and grab a bargain on the components (still very expensive when accounting for TT levers etc), but it'll still be £250 for cables, battery, and junction box. Best case scenario. Given how often I'll use my TT bike, its reasonable to just move the battery over.

    Heres the plan: I bought a GRX di2 LH shifter that has a broken hydro part, di2 is functional. Ill take out the circuit board and the switches. Once its out, essentially all the built in buttons are just momentary switches just like on the sprinter "dummy" port, but the wires for these are soldered onto the circuit. I'll splice into these, and have two on each switch parallel. Shift RD down on basebar and extension, shift RD up on basebar and extension.

    Im planning to 3D print the housings for the switches that go on the basebar and the extensions, and there will be a small box housing the circuit board, Junction A, and the battery that goes in the seatpost. From here, one cable will go to the RD, and 2 more (housing 4-4 wires each) will go to the front of the bike through the downtube. Not planning to have FD on this bike, but if I will, it should work in syncroshift mode so no need for additional buttons for that.

    Ill also have a custom circuit board to eliminate the need of a Junction box B, and cut down on the number of di2 standard wire connections. It will need only 4 actual di2 wire connections (RD, Junction box A, Battery, "shifter" (circuit board from grx shifter).

    Given that I'll do a bunch of circuit testing and my race bike probably wont be complete before next year anyway, I made a board to make testing easier:

    Oh by the way I already have a groupset on it with mechanical shifters. So why this project?

    ...Why not?

  • I'm not sure mounting everything to a board is the most aero solution.

    Is it for testing alignment or something?

  • reverse engineer the canbus protocol

    Something I've been interested in for a while now. Any more info?


Avatar for kiskubai @kiskubai started