Wireless Di2 -- Need to vet ideas

Discuss light weight issues concerning road bikes & parts.
cwolves
Posts: 1
Joined: Thu Feb 02, 2012 12:26 am

by cwolves

I have no experience with Di2 components, but I'm looking into building a wireless Di2 group (probably ultegra) and would appreciate any feedback (especially from an electrical point of view) from anyone that has experience with either of the Di2 groupsets.

First background: I'm getting one of these ( http://www.renovobikes.com/triathlon/ ) in a month and want to build it as clean as humanly possible.

I'm fairly certain that I can bridge the shifters-->derailleurs connection very easily using simple ASK RF communication (prevents cross-talk somewhat like a car remote). From everything I've read the communication between the shifting and the derailleurs, even with the ultegra, is nothing but a high signal passed along.

From the wiring diagram half-way down this page, viewtopic.php?f=10&t=67791&start=285 , I can do away with the junction box entirely. My new plan is:

Code: Select all

RF receiver  <───> Battery 
    ║               │
    ║  Control Box──┼─────────┐
    ║               │         │
    ╚═══════════════╪╦═══════╗│
                    FD       RD


With the RF receiver communicating the RD/FD info only. The FD, RD & Control box are connected to the battery and to one-another via the POS line.

The assumption that I'm making with this, that I can't find any verification on, is that the control box plays no role in controlling the shifting. Specific feedback on this assumption, along with any other feedback is appreciated.

by Weenie


XCProMD
Posts: 584
Joined: Sat Jan 26, 2008 10:25 am
Location: Cantabria

by XCProMD

:popcorn:

kregg
Posts: 143
Joined: Sat Mar 07, 2009 10:04 pm

by kregg

I googled wireless Di2 and came across this. It's not real clear, but maybe hit these guys up.

http://www.youtube.com/watch?v=eEIvv3MBs_k
Tarmac SL3 S-Works 13.75lbs 6.24kg
Jamis Xenith Pro 16.1lbs 7.31kg

Horacio
Posts: 216
Joined: Sat Jan 21, 2012 5:28 am

by Horacio

Saw that video some time ago and it's fantastic! I hope this is just around the corner from Shimano. If not, these guys need to be selling it! Talk about ultra clean.

jeffr
Posts: 518
Joined: Sun Oct 16, 2005 11:50 am
Location: Maui
Contact:

by jeffr

I had two different prototypes working but ultimately I didn't feel it was very practical. And producing a set outside of shimano runs into tremendous patent problems.

To answer the OP, you wouldn't transmit the naked levels of the lines. The transmission frequency would be too great. The POS signal is also used by the front wiring harness/junction to adjust the centering of the rd. So without supporting that you lose trim.

Super_fast
Posts: 341
Joined: Tue Nov 01, 2005 9:36 am

by Super_fast

I made the modification shown in movie above. The bike was shown on the Dutch tradeshow Bikemotion, since we couldn't finish on time for Eurobike and Interbike. The wireless part is built with 2 ultra low power 2.4 GHz transceivers (comparable to Bluetooth and ANT). This wireless transmission is very reliable.

One of the main problems was carbon fiber conducts electricity very well. EM waves (from the 2.4 GHz transceivers) can't get through materials that conduct electricity. Another problem was the current consumption of the receiver. The transceivers consume around 15 mA while sending or receiving. For the sender in the handlebar this is fine, since it can be off until a button was pressed. But the receiver in the frame has to be listening always. We solved both problems, but the solutions weren't straight forward.

With our modification it is possible to adjust the rear derailleur. When you press both buttons of the left shifter for 5 seconds the rear derailleur can be adjusted. Press both buttons again for 5 seconds and it shifts normally again.
Besides the wireless shifting we made it shift faster, and it keeps shifting when you keep a button pressed.

jeffr
Posts: 518
Joined: Sun Oct 16, 2005 11:50 am
Location: Maui
Contact:

by jeffr

Super_fast, whose wireless part did you use? For the trim feature, do you just pulse POS high for a period and then use the normal shift signaling? Or was it more complex than that? I only experimented with this a little bit.

sungod
Posts: 1566
Joined: Fri Dec 24, 2010 9:37 pm
Location: it's raining, it must be uk

by sungod

the signals from my srm and other ant+ sensors to the srm pc7 head unit can get wiped out by speed radars, the effect only lasts for the time i'm in the beam, maybe 100m, but there are roads where it will happen every time

there's the same effect with a garmin 705, although it seems less sensitive to it than the pc7

modern speed radars operate in x/k bands, higher than ant+ (s band), so i assume the signal is just so powerful it's swamping things (although there were old s band speed radars i think they're no longer used)

being unable to shift is not usually going to be a safety issue, but i'd still be wary about relying on wireless for it on the road without plenty of testing

ask rf is 433MHz, so maybe it's going to be less sensitive to radar interference than ant+

one random thought: you'll still have the brake cable to the rear, assuming it is steel, it might be worth experimenting with it as the antenna, maybe with the first length of outer as the other conductor of the feed, 433MHz is c. 700mm wavelength, which could be about the length of brake cable after it emerges from the first outer (btw, i'm not an antenna designer, my first efforts were miserable, after some time i clawed my way to acceptable, but that might've just been luck)

interesting project, hope it goes well!

argus
Posts: 14
Joined: Thu Apr 17, 2014 5:12 am

by argus

Are there any updates to this project? I'm super interested and would be willing to purchase an ultegra di2 to be the guinea pig?

DiscoBoy
Posts: 12
Joined: Wed Feb 05, 2014 3:24 pm

by DiscoBoy

argus wrote:Are there any updates to this project? I'm super interested and would be willing to purchase an ultegra di2 to be the guinea pig?


I doubt that shimano are far off their own version.

Nebby
Posts: 60
Joined: Tue Aug 20, 2013 12:53 am

by Nebby

argus wrote:Are there any updates to this project? I'm super interested and would be willing to purchase an ultegra di2 to be the guinea pig?


The current e-tube protocol would be tricky to figure out without the right equipment, it'd be easier just to wait and see if/when Shimano comes out with a response to Sram's wireless setup.

argus
Posts: 14
Joined: Thu Apr 17, 2014 5:12 am

by argus

Well, I now have my Ultegra Di2's and ordered some parts from SparkFun to see if I can figure it out. Wish me luck!

kulivontot
Posts: 1170
Joined: Sun May 16, 2010 7:28 pm

by kulivontot

If I recall 6770, 6870 and 9070 use a canbus protocol to send data, not a simple high or low, but still should be relatively easy to interface. I'm pretty sure somebody at fairwheel has reverse engineered it. What's kind of battery life are you looking to get out of the shifters/receiver?
I think you'll find that most generic transceivers operate around the 900Mhz or 2.4Ghz range

User avatar
andy2
in the industry
Posts: 713
Joined: Sun Dec 16, 2007 9:33 pm
Location: Luxembourg / Sweden

by andy2

The Shimano e-tube protocol is not a canbus system. It only utilizes two pins, nothing to do with canbus and can not be 'cracked' with canbus analyzers.
rolobikes

kulivontot
Posts: 1170
Joined: Sun May 16, 2010 7:28 pm

by kulivontot

There's multiple threads on di2 hacking. Consensus seems to be that it is some sort of canbus or other serial variant. My point is that it's not a high or low static signal. The easier route may be to get a pair of sprint/climb shifters and hooking your rf transceiver directly into the buttons.

by Weenie


Post Reply
  • Similar Topics
    Replies
    Views
    Last post