Yet Another Di2 Failure

bernithebiker

Some may have seen my previous issues with DA Di2 9070;

1. Front mech defective out of the box
2. Full system crash when installing gear position sender and updating firmware.

Now, I have a new one;

The front mech just stops working. No response to pressing shifter buttons up or down. You are stuck in the big ring (albeit with a functioning rear mech). This has happened twice now.

The first time I resolved it by going into adjustment mode and adjusting the front mech by one notch; this worked and 'revived' it and I was good to go (fortunate as I had another 50km to do).

The second time (today), nothing doing, nothing will revive it, including a full reset. I had to do the last 10km including a 7% hill in the big ring.

Anyone else had this? Any ideas how to fix it?

sloppyschleckonds

Is there a release date for Sram Red wireless?

gimpl

I had the same when my battery was about to run out. It's sort of the equivalent of 'limp mode' in a car. Gets you home.

on-yer-bike

SloppySchleckonds wrote:

Is there a release date for Sram Red wireless?

Will that be different? Unnecessary to run bikes on batteries when they dont need them.

bernithebiker

Gimpl wrote:

I had the same when my battery was about to run out. It's sort of the equivalent of 'limp mode' in a car. Gets you home.

Yes, but each time my battery was 90 and 70%.

But maybe it thinks the battery is low......who knows? Nobody knows, 'cos nobody apart from a few guys in a Japanese lab know how it works......

BrandonA

I've no idea what is causing your problem. Are all your cables attached correctly? I'm wondering if the front mech is loosing power.

Do you have the wireless Di2 add on that lets you see data on your Garmin Edge and also to control the Garmin Edge? I ask if everything is functioning correctly you can view on your Garmin battery life and also the gear you are in. I'd be interested to know what Ant data is being passed to the Garmin. I imagine if the front mech was loosing connection to the system then you might be able to tell in the Garmin.

My wireless adaptor played up one. I unplugged it from the wire and then re-connected. This sort of reset it and fixed stuff. It would take 1 minute to do this with the front mech, might be worth a try.

eddiefiola

i moved my shifters up on the bar a bit the other week and that pulled the cables out of both just enough so that both front and rear didnt shift. pushing them in with a screwdriver (yeah not the right tool) did the job.

bernithebiker

I know the cables are OK, as a) I've checked they are well in position (both at shifter and at front mech) and b) it shifts fine at a later stage. (Having left the bike for an hour or so, the front shifting is back on line.

Yes, I do have the wireless add on, the EWW01. It shows gear position on the Garmin (510) as well as battery level, and also trim position when you go into trim mode.

When the front mech stops working, the front gear position disappears, and it says Position 255. So another sign that it's not a loose cable.

I've asked Shimano what they think (I'm getting to know them quite well), and I'll revert with their answer.

janwal

When I first got my Ultegra Di2 I had intermittent loss of power.Eventually sent it back to bike shop.They traced the fault to the junction box down inside the bottom of the frame by the bottom bracket.They replaced it and 'touchwood' it has run faultlessly since.

964cup

Sounds like an intermittent connection. Is this internal or external cabling? If external, I'd replace junction B; either way I'd also think about replacing the cable from junction B to the front mech. I've had an issue before where having disconnected a cable from junction B, the accumulated dust etc has meant it wouldn't reconnect properly (won't make a solid click home). It still worked, but not reliably.

bernithebiker

964Cup wrote:

Sounds like an intermittent connection. Is this internal or external cabling? If external, I'd replace junction B; either way I'd also think about replacing the cable from junction B to the front mech. I've had an issue before where having disconnected a cable from junction B, the accumulated dust etc has meant it wouldn't reconnect properly (won't make a solid click home). It still worked, but not reliably.

It's all internal, with internal battery. Thing is, I haven't moved or disconnected anything, so no reason for a cable to be loose or have a dirty connection. But I guess maybe I should disconnect everything, spray in some cleaner and reconnect.

(I hardly ever ride in the rain either, and the bike is immaculate).

snakedoc

Sorry to hear you've had so many issues with the Di2 Bernie, I know you sat on the fence awhile before deciding to go for it so even more frustrating.
Mine has been fine (touch wood) - a real case of bad luck.
Hope it's gets sorted soon

964cup

bernithebiker wrote:

964Cup wrote:

Sounds like an intermittent connection. Is this internal or external cabling? If external, I'd replace junction B; either way I'd also think about replacing the cable from junction B to the front mech. I've had an issue before where having disconnected a cable from junction B, the accumulated dust etc has meant it wouldn't reconnect properly (won't make a solid click home). It still worked, but not reliably.

It's all internal, with internal battery. Thing is, I haven't moved or disconnected anything, so no reason for a cable to be loose or have a dirty connection. But I guess maybe I should disconnect everything, spray in some cleaner and reconnect.

(I hardly ever ride in the rain either, and the bike is immaculate).

Could be an issue from the original installation, I suppose. Well worth redoing all the connections, checking for a proper click; you could use dielectric grease as well to ensure waterproofing. If your LBS has the full diagnostic kit, it's worth using that to run some test cycles.

meanredspider

All this makes me wonder if you have an undiagnosed fault somewhere that is showing up in various places with different symptoms. Two Di2 bikes, several years, thousands of kms and not a single glitch and less than 10 minutes of fettling (mostly due to a damaged mech hanger after a crash).

First.Aspect

There's nothing you've said which isn't consistent with a loose connection somewhere. - Things work fine sometimes. - Intermittent fault. - Just done some fettling.

Could be a duff box, but start with the easy stuff. Otherwise you are just being stubborn.

bernithebiker

Shimano said to ;

a) check the cable connection at the shifter. OK fine, but the shifter is well tightened to the bar, and I left some slack in the cable so no reason for this to be loose.

b) interrogate system with the charger system (that I don't have, it's the other one) and update all firmware (which I did when I had the last problem).

So will try those 2 things on Tuesday when I get back from the Mallorca 312 (where fortunately I keep my 2nd bike, non Di2!!!)

964cup

Bear in mind you might have a faulty cable - if you have a dry break, for instance, it will connect most of the time, but unreliably. The same would be true of a connector that's not fully seated. I've had four Di2 bikes (I have two at present) and have >7.5k km on them with no issues that were not of my own making, so I'd be more inclined to look for a physical or installation fault with your particular bike than blame this on the system. I've not personally had a dry break cable in Di2 (although I've had to diagnose this kind of fault in lots of other electrical systems), but I have had the connector-not-fully-home problem, notably with external junction B and with an FD where I tried an alternative cable route that avoided the cable strain relief clip on the back of the FD.

ugo.santalucia

meanredspider wrote:

All this makes me wonder if you have an undiagnosed fault somewhere that is showing up in various places with different symptoms.

... such as diabetes? :shock:

bernithebiker

964Cup wrote:

Bear in mind you might have a faulty cable - if you have a dry break, for instance, it will connect most of the time, but unreliably. The same would be true of a connector that's not fully seated. I've had four Di2 bikes (I have two at present) and have >7.5k km on them with no issues that were not of my own making, so I'd be more inclined to look for a physical or installation fault with your particular bike than blame this on the system. I've not personally had a dry break cable in Di2 (although I've had to diagnose this kind of fault in lots of other electrical systems), but I have had the connector-not-fully-home problem, notably with external junction B and with an FD where I tried an alternative cable route that avoided the cable strain relief clip on the back of the FD.

Noted, but the thing is, the bike has worked fine since installation about 6 months ago (apart from the EWW01 problem, 3 months ago which I sorted)

So about 1250km per month with no issues, then suddenly this new problem; but the bike has not been dropped, soaked or abused in any way. I know I pushed the connectors in firmly when assembling, (you get that loud click when they go in properly, and they're hard to pull back out) so no reason they should have worked loose.

Anyway, I'll go through them one by one when I get the chance. Maybe a wire has chafed through somewhere?

Grill

Strange one. Could be a faulty junction box or cable...

Mine has been flawless even though it's been fully submerged and using 5 junction boxes...

First.Aspect

bernithebiker wrote:

964Cup wrote:

Bear in mind you might have a faulty cable - if you have a dry break, for instance, it will connect most of the time, but unreliably. The same would be true of a connector that's not fully seated. I've had four Di2 bikes (I have two at present) and have >7.5k km on them with no issues that were not of my own making, so I'd be more inclined to look for a physical or installation fault with your particular bike than blame this on the system. I've not personally had a dry break cable in Di2 (although I've had to diagnose this kind of fault in lots of other electrical systems), but I have had the connector-not-fully-home problem, notably with external junction B and with an FD where I tried an alternative cable route that avoided the cable strain relief clip on the back of the FD.

Noted, but the thing is, the bike has worked fine since installation about 6 months ago (apart from the EWW01 problem, 3 months ago which I sorted)

So about 1250km per month with no issues, then suddenly this new problem; but the bike has not been dropped, soaked or abused in any way. I know I pushed the connectors in firmly when assembling, (you get that loud click when they go in properly, and they're hard to pull back out) so no reason they should have worked loose.

Anyway, I'll go through them one by one when I get the chance. Maybe a wire has chafed through somewhere?

Your argument is based on the misapprehension that everything that used to be okay must still be okay because you've not done anything to make it not okay.

Things break or work loose when they are shaken violently up and down for a few hundred hours. Something might have worked loose, so why not take peoples' advice and check systematically?

If you can't find the issue and take it to the lbs, the first thing your kind mechanic will do is check all of the connections. How much of a d!ck are you going to feel if he pops one back in and gives you a call before you've managed to make it home? We'd expect a follow up post on the forum, of course.

meanredspider

It's very easy to check the shifter connection. Both of mine were loose (LBS guy who assembled it - I was too busy - hadn't done Di2 before) and only showed up after time. Both easily fixed mid-ride. If it was working then something has changed (to state the obvious) - you just need to find it and do so in a logical way. After many many years managing maintenance departments and engineering projects, I frequently experienced people jumping to conclusions or trying to fix the wrong problem and I never once experienced a fix that didn't make perfect sense.

bernithebiker

OK, OK, I am going to check the connections (like I said!) I just can't do it right now as I'm not at home!!

Maybe it is that. But maybe it could be several other things;

a) shifter or derailleur failure
b) junction box failure
c) chafed wire
d) confused communication with battery

We shall see! If I find the problem, I will post it here, whether it makes me look stupid or not..!

sloppyschleckonds

Damn, looks like 2016 release

http://www.cyclingnews.com/news/pozzovi ... sram-gears

964cup

bernithebiker wrote:

c) chafed wire

ISTR you're on internal cabling. If your BB isn't sleeved then chafing against the crank spindle is a definite possibility, so don't omit pulling the crank to check once you've eliminated everything else. Depending on the frame routing you may also need to check inside the headtube (e.g. Giant, Cervelo S5 etc) as the cable enters the toptube, routes through the headtube and then into the downtube. Doubt the steerer will chafe it as such, but it could pinch, I suppose - although I would expect that to affect both derailleurs, obviously.

bernithebiker

964Cup wrote:

bernithebiker wrote:

c) chafed wire

ISTR you're on internal cabling. If your BB isn't sleeved then chafing against the crank spindle is a definite possibility, so don't omit pulling the crank to check once you've eliminated everything else. Depending on the frame routing you may also need to check inside the headtube (e.g. Giant, Cervelo S5 etc) as the cable enters the toptube, routes through the headtube and then into the downtube. Doubt the steerer will chafe it as such, but it could pinch, I suppose - although I would expect that to affect both derailleurs, obviously.

Good points, but the BB is sleeved on the Tarmac (was a pain to get the rear mech cable over and round it!).

And yes, cable does enter top tube, goes round head tube, and down down tube, so worth inspecting here too.

mtb_mark_uk

I'm not familiar with the Di2 setup but I do know a thing or two about electronics (and have built my own power amps from scratch). The easiest way to test this would be expose the connectors at each end of the cable (if possible with Di2?) which you suspect is at fault. Then use a multimeter to do a continuity test across the ends of each of the signal/power wires to ensure there's no breaks. Assuming the cable is fine, and the front mech still doesn't work then you will have to turn your attention to other parts of the system.

As has been said before, it sounds like your fault is most likely caused by vibration induced dry joints somewhere. ie/ This is where the solder has lost it's metallic bond to a wire and electrical contact is only maintained by metal to metal pressure only (increasing the resistance). A vibration in the correct place could easily cause them to move physically apart and hence lose the connection.

raldat

I truely am not trying to be rude or condescending but you have said on a number of occasions that it worked fine in the past and that nothing has changed etc. Unfortunately with electronics this is not how it works. (Obviously) electronics work by sending little messages as electrical pulses. All it takes for that to get through is a tiny connecting between two copper wires or connectors. Something with a break in it may work just fine for a long time if that connection is maintained, then something changes. It might be as minor as a small bump or a little corrosion setting in on a connector but if such a fault exists, that is all it takes to break the connection. If the copper strands on a cable are broken, they may be touching and work, but if a small gap appears, the message no longer gets through. This could me mere microns of movement. These types of faults can come and go for an extended period until at some stage, something will shift enough to break the link permanently.

The only solution to this is to methodically work through all connections and cables until the fault is isolated. That is the problem with a reliance on electronics. I am not criticing your choice and I am no luddite, but a switch from mechanical to electronic requires a very different set of trouble shooting techniques.

The most common cause of these issues is corrosion on connectors. Sometimes pulling connectors and reconnecting them is all that is needed to fix the issue as that act can clean the corrosion.

Good luck with your quest.

bernithebiker

Have unclipped some of the wires connected in some way to the front mech. All were clipped in fine, all seem clean and intact.

Finally got to do a 60km hilly ride, plenty of front shifts; absolutely fine, no issues.

But I had about 300km of no issue riding between the two failures, so I'm not confident the problem won't come back.

I was wondering; as both failures were when in the big ring, if it fails again in the big ring, isn't that too coincidental? i.e. points to something other than connection issues?

First.Aspect

So, what you are saying is that you reconnected everything and now its working?

Well blow me down.

bernithebiker

First Aspect wrote:

So, what you are saying is that you reconnected everything and now its working?

Well blow me down.

Maybe you didn't read the whole thread? There was a BIG gap in between the two failures (couple of weeks, 3 to 400km). So it's reasonable to assume there could be a big gap before the next one, no?

I've only done 60k on it so far, and I'm far from confident the problem has gone away. We shall see.

In any case, given that the connections and wires seem absolutely fine, if that was it, then isn't it a bit worrying from a long term reliability stand point?