Di2 Difficult Diagnosis

[Bollo]

I've a 9070 DuraAce Di2 setup on my bestest bike. It's been fine for 18 months but has now developed a very strange and difficult to diagnose intermittent problem. I'm putting this out there to see if anyone has had the same problem and, if it gets fixed, as a record of WTF caused it.

Here's as far as I've got. Apologies for the length but it's all relevant, promise!

Seemingly at random, the entire shifting system shuts down - front and rear derailleurs both dead. Pressing both shifters would normally flash the junction box lights to show battery level, but this also appears dead. Pressing the junction reset button does nothing.

This next bit is where it gets weirder. For most rides, the whole system recovers after between 5-20 minutes. It just all comes back. However, on my last ride I'd removed the D-Fly transmitter and, sure enough, after half an hour everything stops shifting. Except this time it doesn't recover and I have to make it home in a vast gear. With the D-Fly attached, the system comes back up showing adjustment mode on my Garmin, just like this well-known and still open issue https://forums.garmin.com/archive/index.php/t-90282.html.

Also, on the two occasions I've made it back with the system still down, it immediately comes back the moment the junction box is plugged back in to either a charger or a computer. After that point, all is well again, even if the charger cable is immediately removed.

Finally, the D-Fly still operates my Garmin through the hood buttons when the system is down, suggesting that there's still power coming from the battery.

In terms of fixes, I've made sure the battery is charged and removed and reseated all connectors except the internal battery, which is a bugger to get to.

My current money is on a faulty junction box. All the cables looked in good nick, the battery is nowhere near its life and is well protected in the frame. Also, the way the system shuts down and then restarts and the active D-Fly point away from a battery issue, and it's immediate recovery on plugging in wouldn't be typical behaviour for a cable or connector fault.

My TT bike also has Di2, so I'll swap out the junction box from there and see if that makes the problem go away on tomorrow's ride.

Anyone got any other ideas?

 

[dan_bo]

Set fire to it and get cabled sora.

 

[Bollo]

Set fire to it and get cabled sora.

Cables?!

SORA!!?!?!

 

[avsd]

Loss connection on the internal battery but try the swap of the junction box first.

 

[derrick]

I've a 9070 DuraAce Di2 setup on my bestest bike. It's been fine for 18 months but has now developed a very strange and difficult to diagnose intermittent problem. I'm putting this out there to see if anyone has had the same problem and, if it gets fixed, as a record of WTF caused it.

Here's as far as I've got. Apologies for the length but it's all relevant, promise!

Seemingly at random, the entire shifting system shuts down - front and rear derailleurs both dead. Pressing both shifters would normally flash the junction box lights to show battery level, but this also appears dead. Pressing the junction reset button does nothing.

This next bit is where it gets weirder. For most rides, the whole system recovers after between 5-20 minutes. It just all comes back. However, on my last ride I'd removed the D-Fly transmitter and, sure enough, after half an hour everything stops shifting. Except this time it doesn't recover and I have to make it home in a vast gear. With the D-Fly attached, the system comes back up showing adjustment mode on my Garmin, just like this well-known and still open issue https://forums.garmin.com/archive/index.php/t-90282.html.

Also, on the two occasions I've made it back with the system still down, it immediately comes back the moment the junction box is plugged back in to either a charger or a computer. After that point, all is well again, even if the charger cable is immediately removed.

Finally, the D-Fly still operates my Garmin through the hood buttons when the system is down, suggesting that there's still power coming from the battery.

In terms of fixes, I've made sure the battery is charged and removed and reseated all connectors except the internal battery, which is a bugger to get to.

My current money is on a faulty junction box. All the cables looked in good nick, the battery is nowhere near its life and is well protected in the frame. Also, the way the system shuts down and then restarts and the active D-Fly point away from a battery issue, and it's immediate recovery on plugging in wouldn't be typical behaviour for a cable or connector fault.

My TT bike also has Di2, so I'll swap out the junction box from there and see if that makes the problem go away on tomorrow's ride.

Anyone got any other ideas?

Which junction box have you looked at? There are two.

 

[Bollo]

Loss connection on the internal battery but try the swap of the junction box first.

As I said, the system comes back immediately the junction box is connected to the charger d the D-fly remains functional, so not a battery connection. I've reproduced this twice. Given the cable routing from battery to junction box, it's unlikely that I've jiggled the battery connection at the exact moment I've plugged in the charger. The charger only needs to be plugged in for a second and the whole thing comes up and stays up, so not a loss of charge. It's like the system is being rebooted.

 

[Bollo]

Which junction box have you looked at? There are two.

3 port A junction. The bike is a Canyon Aeroad so it's mounted in a recess in the integrated stem. This area can pick up spray and crap, so that's a reason to suspect. The B Junction is internal so a bit of a faff to get to. I'm only going after that if the A junction theory fails. Only a sense at the moment, but the behaviour feels like an electronics/software issue over something purely physical like a crap connection. I've not updated the firmware recently either.

 

[dan_bo]

Cables?!
View attachment 337385

SORA!!?!?!
View attachment 337386

haha excellent.

 

[User33236]

Assuming it's a single fault condition it seems, from your description, that the 'front end' isn't talking to the rest of the system to likely to one of three components.

Junction box A
Junction box B
E tube cable between A & B

As you already plan to do, swap out each item in order of easiest first, which is likely to be as ordered above, and see how it goes.

 

[Bollo]

@sre! Exactly half an hour into today's ride with the alternative A Junction from the TT bike fitted and it all goes south again. This time a twiddle back home in a climbing gear. As before, it recovered after about 25 minutes, but then went off and on again once more over the course of a minute or so.. This would be much more like a connection issue. It was still working when I got home so no opportunity for further diagnosis.

Like the other times, on this occasion the D-Fly remained active so I think this narrows it down to either the connection somewhere between the two junction boxes and the battery or some bizarre battery issue that keeps the D-Fly powered but shuts down everything else.

Getting at the battery and internal cables requires a BB92 removal tool, which I do not have. As the BB is in pretty tight I'm not going to improvise, so it'll be a couple of days until I can fish the battery and cable out.

Thanks for the suggestions so far. Have likes.

 

[Dirk]

Sounds like it's possibly a faulty battery, with the voltage collapsing under load.
The rear mech would pull much more load (current) than the D-fly as it is using servos.
The battery would still be able to provide the power for something like the D-fly as it would only be drawing an extremely small current.

 

[Bollo]

Sounds like it's possibly a faulty battery, with the voltage collapsing under load.
The rear mech would pull much more load (current) than the D-fly as it is using servos.
The battery would still be able to provide the power for something like the D-fly as it would only be drawing an extremely small current.

I can see the reasoning but it's the way it recovers completely after a dead period that has me mystified. One thing I can't find out is how much 'logic' (electronics, firmware) lives on the battery. While it's unlikely the chemical side of the battery would give an intermittent fault with these chararctistics, its possible that any attached chips etc might. I still need to look for the more obvious/cheaper stuff first though, so internal cabling out tomorrow.....

 

[User33236]

There is a fair bit of electronics and programming within the battery which seems to be the brains behind the system. Indeed to use some of the latest features such as synchro-shift. the system requires the latest battery with its increased memory.

 

[Bollo]

There is a fair bit of electronics and programming within the battery which seems to be the brains behind the system. Indeed to use some of the latest features such as synchro-shift. the system requires the latest battery with its increased memory.

Reading around I was getting that impression, but I just wish the Shimano website was in any way useful.

Slightly OT, but they've just set up a High Tech Campus in Eindhoven, NL, about half a mile from some friends that we often visit. I might pop in next time (and see if there are any jobs ;-P)

 

[SheilaH]

Anyone got any other ideas?

Yes.

Take it to a bikeshop who have the diagnosis kit.

You've 3 years warranty, and you should make use of it. You've paid for it.