I have ETAP on two bikes, one is symptomatic, the other is not.

The symptom is that the rear derailleur stops responding / shifting (mid-ride) as if the battery were dead. I initially thought the problem was that it was draining the rear battery while parked... but "wait and watch" has indicated that the batteries are not dead, just that electronics go flaky, oddly, starting out fine, but flaking out while actually out riding.

The resolutions to date have been freakishly random.

One time, I shifted the front down then back up, just to see if the front still worked... and then the back started working again. WTF? Usually however shifting the front does not reset it.

One time I swapped front and rear batteries and both started working again.

One time I removed the back battery and simply put it right back on and it started working again.

One time I swapped front and rear batteries and the rear started working and the front stopped. Oops. My bad. That was just a dead battery lol.

One time I swapped front and rear batteries, and the front continued working and the rear did not. As I was heading into some serious hills, I turned around to head back. I dunno why, but I stopped and swapped the batteries again, just for perverse kicks... and it magically started working again! So I turned back around again and finished my ride.

It has never happened (so far at least) twice on the same ride.

It seems very much like a software crash to me. Anyone ever hear of anything like this?

I did end up getting it from a LBS so I have full US warranty. I'll drop it at the shop and make it their problem eventually.

But I'm really hoping there is either a software fix or a warranty replacement that I can do myself (as shop trips eat up mega free time for me).

My stock answer is that you don't need a mechanic, you need an electrician. These days I might need to update it and say you need a computer programmer.

My one and only beef about electronic shifting is that it's not amenable to problem solving by observation and analysis. Something totally invisible, like a sloppy internal connection, a program glitch, or whatever can be the issue, but there's no way to tell.

My SOP for stuff like this, on all electronics, is to power down, remove the battery, let it rest for long enough for any residual charges to drain or equalize. Then power back up and hope for the best. Sometimes I'm lucky, and sometimes I restrain my impulses to chuck it out a window.

Kind of sums up why I'm hesitant to go electronic on the bike. With the mechanical it's easy enough to see why something isn't work and either fix it or replace it. Electronic issue not so much.

For the record (and IMO) eTap is definitely worth working through this issue.

While there is absolutely nothing about etap that is functionally critical or that I can't live without, aesthetically it is so much nicer that I just don't want to live without it.

Being a programmer doesn't help. There's (human understandable) source code, but when a product goes to market, that gets translated into machine code, and then we don't know what it says.

It's a point, however facetious comments don't lend themselves well to serious analysis.

Holy crap that was a fast reply.

On the net, you see a lot of public opinions about what programmers do, it kind of gives you the idea that most folks thing programmers are like anti-social wizards who can do anything with technology. We don't get to choose what we work on, and we take our Garmins and eTaps back to the store to complain just like anybody else.

Just be sure to leave it in the non-working condition when you take it in to the shop. I know that when out on a ride you will be tempted to get it running again, but you have got to demonstrate a problem, or they can't do anything for you warranty-wise.

Seems two areas of investigation. First is the comms channel between the shifter and the DR. Second is the internal logic in the shifter and DR. Since the comms have to be generated by both shifters, first thing would be if it's upshift or downshift (I assume both or OP would have mentioned it). If that's correct then next would seem to be the receiver in the rear DR. It would be nice if you could remove a shifter and do some test shifts in different ranges or use some blocking to see if you can get the DR to stop responding.

As an aside, do the shifters "know" what cog the rear is in? That is, if the rear is all the way left or right do the shifters still try to shift farther or not?

scott s.
.

scott: Once they start working again, they know where the back knows where it is and doesn't don't try shifting past the limit screws. And the front knows what ring it's on too.

rpenman - Good point. Now that I've decided it needs to be dealt with, I will at least take a video of it non-working and document how I get it working again.

Preface: I have no direct knowledge of etap other than what I've read. Have you checked and/or replaced the batteries in the shifters? Just a thought.

Since swapping batteries around seems to frequently fix the issue I'd start by doing a thorough cleaning of all contacts on the batteries themselves and the derailleurs. As I've found with my Nikon flashes a tiny bit of oxidation on the contacts can result in zero power to the device and/or lots of fruitless battery swaps. Use some electronics cleaner and a q-tip or a pencil eraser depending on what you have handy.

Also check that there are no cracks in any of the retaining clips or whatever SRAM uses to attach the batteries. I have heard of issues of the hooks of the batteries breaking off and perhaps yours are just to fail.

I will try cleaning the contacts on the derailleurs and on all batteries and see if it recurs. Clips are in good condition, but I will inspect.

Did not change Brifter 2032's as the problem seems very independent of them.

Didn't read the specifics on everything but noticed mention of battery issues - the only issue I keep running into with etap at races are riders saying they are having issues with the batteries. Either not working, holding a charge, etc. Usually when they approach SRAM race support they are simply handed a new battery and that's the end of it.

Smacks of a known issue.

Don't know if that helps or not.

You know, a lot of cr2032 batteries are old, fake, or not holding a full charge.

Your safest supplier for legit cr2032 at a good price is Digikey.

Slightly OT, but part of me believes that it's only a matter of time before people who grew up in the era of electronic shifting -- young'ns who don't have the years of experience trouble-shooting mechanical drivetrains but also never knew a world that didn't have SmartPhones, X-box, a home PC and WiFi -- will outnumber us, and for those (now) kids this sort of problem solving by observation and analysis for electronics will be second nature.

iow, every time I hear someone insist that they won't be getting eTap or Di2 because they can't imagine troubleshooting electronics, I think that's a failure of the imagination rather than an inherent flaw of electronic shifting.

Doesn't seem a battery (etap or 2032) issue, as moving etap batteries around always fixes the issue, and it only happens on one bike, never the other.

I don't have much experience with eTap (Di2 guy), but isn't there a function button on the RD that allows you shift inboard or outboard with a single or double-click? When the RD is acting up, are you still able to shift using the function button on the RD? If so, it may be a pairing issue with your shifter, and the RD is fine. If you can't shift with the function button, then the problem is narrowed down to something inside the RD (like a sketchy servo).

Thx. Next time it acts up i will also try the function button. When it acts up, it wn't shift up or down, so getting it to move one direction will likely get it working for both. Or not, if the crash is on the wireless comm side of the interface. Still a useful data point.

Makes me think removing the battery erases volatile memory and then it "reboots".

scott s.
.

Yeah. Except for the time that just shifting the front fixed it. Or the time that removing the battery didn't fix it. Or at least it took two shots at it. :shrug:

I (re)cleaned all the contacts on both biks, all 4 def, all 4 batteries.

Waiting for it to recur for video.

Have you tried swapping batteries from one bike to the other?

Battery to derailleur paring is random, as I charge batteries and move them around randomly w/ no real pattern. Example, heading out for a ride, taking bike #1, but not sure when charged last. I'll grab batteries from bike 2, and put these on charge. or maybe bike 2 has no batteries, so I'll take the two off charge and put on bike, and put these on charge. But charger to bike is random. And bike to bike, I'll intentionally swap front to back on the assumption that the back always gets more use.

Problem has stayed w/ Rear Der on Bike #1, and has not appeared anywhere else.

Moving batteries fixing it...says to me the problem is how the batteries are seating, or more precisely are not seating properly on the contacts.

Not having an ETap system anywhere in my entire state to even look at one IRL probably, I can only guess at your issue. Have you asked SRAM?

I still vote that moving the batteries is just resetting the system and the batteries themselves are not at issue. Loss of connectivity between shifter and RD seems a possible failure mode. "Waking up" on a front shift may mean that there was a state transition that re-synced the RD. (I assume the RD receives both L/R shifter signals simultaneously and then "ignores" them and so doing might put the RD into a certain state transition, further assuming there is some sort of finite state machine that implements the comms protocol).

Do you ever try a simultaneous shift? IIUC you can operate both shift levers, then release one while holding the other to get "near-simultaneous" shifts. Or vice-versa.

Agree that trying the manual on-DR control when it isn't responding to shifter gives another data point.

scott s.
.