Assembly of Di2

On Thursday, January 30, 2020 at 7:30:17 AM UTC-8, jbeattie wrote:
On Wednesday, January 29, 2020 at 9:39:54 PM UTC-8, Tom Kunich wrote:
On Wednesday, January 29, 2020 at 6:02:32 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 1:52:43 AM UTC, jbeattie wrote:
On Wednesday, January 29, 2020 at 3:52:41 PM UTC-8, Andre Jute wrote:
On Wednesday, January 29, 2020 at 8:26:32 PM UTC, Tom Kunich wrote:
On Wednesday, January 29, 2020 at 12:00:57 PM UTC-8, jbeattie wrote:
On Wednesday, January 29, 2020 at 11:17:16 AM UTC-8, Tom Kunich wrote:
On Tuesday, January 28, 2020 at 5:36:21 PM UTC-8, jbeattie wrote:
On Tuesday, January 28, 2020 at 4:59:05 PM UTC-8, Tom Kunich wrote:
On Tuesday, January 28, 2020 at 1:40:42 AM UTC-8, wrote:
On Monday, January 27, 2020 at 1:59:08 AM UTC+1, Tom Kunich wrote:
On Thursday, January 9, 2020 at 9:43:42 AM UTC-8, Tom Kunich wrote:
It occurs to me that I should install the hydraulic brakes before I attach the electronic wiring. What do you think?

Well, I have put the Di2 together. And nothing works. The possible sources of error a

1. The wiring is incorrect. I assumed that the left and right shifters would go into the left and right connector of to two holes of the 5 hole Stem unit. The manual is not clear about this so it was ad lib.

2. The new battery could be flat. I have no way of measuring the voltage under load but the unloaded voltage is 8 volts and Shimano says that 7 something volts is a full charge. I bought a new charger but I don't think it was new since it didn't come in a box or sealed plastic bag and when I questioned the seller he promptly returned my money and said that I could keep the charger. Looking at the output voltage it measures zero but I don't know what the charging circuitry is. It could turn off with no detectable load. Hmmm. So I bought another "new" charger and hopefully this one will work.

So the question is this: Could the battery be so flat that the unit wouldn't turn on?

Possible, but I never came across a new battery out of the box that was so dead that it wasn't able to power a led.

Lou

I find it curious as well. Nut Lithium Ion batteries are odd works. Perhaps plugging it into the battery charger whish isn't working drained the battery. And the high voltage measurement I got was because there was no load on it.

In any case the new charger should arrive Friday and I'll test the charge again.

Otherwise I'll have to replace the stem unit. I have no idea why his stem junction box is a five port unless he was running TT bars with end shifters.

Is this a trick question? If your battery is showing 8v, it's not dead. No? Why would load matter -- except for measuring capacity.. I understand the battery may have a protection circuit, but wouldn't that decrease the measured voltage?

I don't have an EE degree, but I measure batteries with a multimeter all the time to see if they are alive or dead, and if one showed 8v it would not be dead. I think you messed something up. Just check the quality of your connections. Do you have the little tool? https://www.youtube.com/watch?v=OkdY9q-u5Dk If you're not fishing wires, it's really pretty fast and easy. Its the fishing that takes time.

-- Jay Beattie.

Jay, stop showing that you don't know anything about engineering. A battery is similar to a capacitor to the extend that load matters to output voltage. Even a small charge on a battery can show "normal" voltage if you are measuring it unloaded.

So what is the likelihood that a battery showing 8v will not produce enough current to run a tiny LED? I mean your not cranking a starter motor. I represent these guys and will see what they have to say. https://mobilepowersolutions.com/ Every time you have some problem, it turns out to be some bizarre mistake, as with your bottom bracket measurement, seat tube clamp diameter, etc., etc. This is not rocket science. Di2 is an easy install.

-- Jay Beattie.

Tell us how many Di2 you personally installed? What is bizarre is that you represent some people and hence know electronics engineering. Or that you believe that the only thing inside that stem junction box is an LED.

Jay would be very ****ed off if those three electronics engineers went around pontificating about the law because "we know Jay Beattie, who's a lawyer".

I agree, and my disbelief is based on my experience with the product and not superior electrical knowledge. I totally shorted a Di2 battery (sparks and all when I cut the rear derailleur wire), and it still had enough juice to run the flashing red indicator LED on the junction box. It wouldn't shift a gear, but it would run the light.

I understand that the voltage difference can be minimal between a charged and discharged battery -- maybe only a volt or two, but Tom said the measured output was 8v. The Di2 battery is 7.4v battery and charges up to a little over 8v, so it would appear he has a NEW, charged battery. He would then have to have a battery that couldn't manage even a tiny load -- the same load handled by my shorted battery. I understand that can happen, but what is the likelihood compared to a screw up in connection, bad junction, etc., etc. My multimeter has a battery load tester, so I would just use that anyway. Tom should get one of these. $10 at Home Depot. https://tinyurl..com/qsj599e My money is still on some other issue.

-- Jay Beattie.

My money is on the wiring, not because I know the wiring chart on Tom's Di2 for road bikes off by heart, but because on my much more complicated commuter's fully auto plus adaptive suspension Di2 when I reengineered it to suit me (and the brief given to the designer who had delusions of sporting activities) better, extending the wiring was the thing that gave me the most bother even though it was supposedly a plug-and-play system and it worked perfectly when it arrived so that I in fact had a working blueprint to work on.

Andre Jute
Working with 1500V for my tube hi-fi was less fraught than working with Di2

The stem junction box is the system controller. It has a microprocessor in it to generate the system control language. All it requires is power to make the LED's light. So all it requires is a power line. Since I ran a system wire directly from the battery to the stem control unit and it didn't light there can only be two sources of error. The battery is too low or the stem unit is defective. I am taking the seller's word that it worked fine and I know that my battery charger is defective so it must have be the battery charge.

The new charger is due on Friday so Saturday I'll know. Also I am on the verge of buying a three connector stem unit since that is what was supposed to come on the group. If all else fails replace the only two sources of error.

Also there are only two types of wiring - the levers which only turn the voltage on frontwards or backwards to shift up or down and ALL of the rest of the wires are common so as long as they are plugged in they are correct. In short, it is nearly impossible to wire it wrong.

I hope that works for you. I just downloaded the e-tube project software and will update the firmware on all my components this weekend and maybe do some customization. I don't like the auto rear downshift shift when I downshift the FD, and I want to see if I can disarm the anti-cross chaining feature. My recharging cord works as a link, but you may need to buy the e-tube PC linkage device. Anyway, I haven't done this yet, but if you have a link and he downloadable etube project software, it will run diagnostics on your system, including the battery.

-- Jay Beattie.

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

On 30/01/2020 16:36, Tom Kunich wrote:
On Wednesday, January 29, 2020 at 10:49:56 PM UTC-8, Tosspot wrote:
On 30/01/2020 00:36, Tom Kunich wrote:
Jay, an LED requires 1.5 volts bias or so to turn on. This is why
I was saying to you to not try to be an EE. You can easily
discharge the battery to below the bias simply by leaving
everything on. At $100 for a battery I sure hope this bad battery
charger didn't ruin the battery.

I'm pretty sure the Di2 contains a couple of 14430s, at about
10USB. I'd have also thought an LED would start conducting around
the 0.7V mark. Ok, not very bright. Btw, I'm not an EE either.

There is a difference between forward bias voltage of a transistor or
common diode and an LED:

Forward-bias is when the anode (the pointy part of the symbol) is
positive and the cathode (the bar) is negative. Reverse-bias is when
the anode is negative and the cathode is positive. ... Typically, the
forward voltage of an LED is about 1.8–3.3 volts; it varies by the
color of the LED. A red LED typically drops 1.8 volts, but voltage
drop normally rises as the light frequency increases, so a blue LED
may drop around 3.3 volts.


You are clearly more of an EE than me folds

On Thursday, January 30, 2020 at 11:03:59 AM UTC, Duane wrote:
Andre Jute wrote:
On Wednesday, January 29, 2020 at 6:07:40 PM UTC, Duane wrote:
On 1/29/2020 10:05 AM, jbeattie wrote:
On Tuesday, January 28, 2020 at 11:01:22 PM UTC-8, Tosspot wrote:
On 29/01/2020 02:36, jbeattie wrote:

snip

Is this a trick question? If your battery is showing 8v, it's not
dead. No? Why would load matter -- except for measuring capacity. I
understand the battery may have a protection circuit, but wouldn't
that decrease the measured voltage?

I don't have an EE degree, but I measure batteries with a multimeter
all the time to see if they are alive or dead, and if one showed 8v
it would not be dead. I think you messed something up. Just check
the quality of your connections. Do you have the little tool?
https://www.youtube.com/watch?v=OkdY9q-u5Dk If you're not fishing
wires, it's really pretty fast and easy. Its the fishing that takes
time.

A battery will often show volts but not be able to pwoer anything. You
should measure the voltage in circuit if possible (i.e. under load), or
just check the voltage of an old PP9 you have lying about (smoke
detector battery). If you have a *lot* lying around.

https://www.youtube.com/watch?v=8hwLHdBTQ7s

I would think that if the battery is showing full voltage, it would
have enough amperage to run an LED at the junction box unless there was
something really weird about the protection circuit, but I'll leave
that to the electronic gurus. With Tom, it always turns out to be
something mundane like a wrong part or a mis-description, like his very
unique BB which turned out to be a standard 68mm ISO BB. The project
should take an hour or so and not months.

-- Jay Beattie.


Irrespective of Tom's abilities, it's not uncommon for a battery to show
voltage without a load and be dead.

More precisely, to die the moment a load is put on it, and to resurrect
itself after the load is removed, hence the universal consciousness of
rising on the third day.

Andre Jute
Hallelujah!



I prefer the Schrodinger cat analogy. The battery is dead and not dead
until you connect a load to it.

My wife keeps cats. I dare not speak of a dead cats. -- AJ

On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

On Thursday, January 30, 2020 at 11:32:15 AM UTC-8, Tosspot wrote:
On 30/01/2020 16:36, Tom Kunich wrote:
On Wednesday, January 29, 2020 at 10:49:56 PM UTC-8, Tosspot wrote:
On 30/01/2020 00:36, Tom Kunich wrote:
Jay, an LED requires 1.5 volts bias or so to turn on. This is why
I was saying to you to not try to be an EE. You can easily
discharge the battery to below the bias simply by leaving
everything on. At $100 for a battery I sure hope this bad battery
charger didn't ruin the battery.

I'm pretty sure the Di2 contains a couple of 14430s, at about
10USB. I'd have also thought an LED would start conducting around
the 0.7V mark. Ok, not very bright. Btw, I'm not an EE either.

There is a difference between forward bias voltage of a transistor or
common diode and an LED:

Forward-bias is when the anode (the pointy part of the symbol) is
positive and the cathode (the bar) is negative. Reverse-bias is when
the anode is negative and the cathode is positive. ... Typically, the
forward voltage of an LED is about 1.8–3.3 volts; it varies by the
color of the LED. A red LED typically drops 1.8 volts, but voltage
drop normally rises as the light frequency increases, so a blue LED
may drop around 3.3 volts.


You are clearly more of an EE than me folds

When I say that I'm an EE I'm not kidding. I worked on large projects all over the bay area which at the time was mostly Silicon Valley. I had so many engineering awards that I got tired of them and threw them out. I completed projects that PhDs said couldn't be done. (whether or not they really meant that or were trying to get more research grant money is a question.) I can't even count the number of research breakthroughs I made.

Here's one I missed - blah, blah, blah - "Thank you for all the knowledge, creativity and commitment you have dedicated towards the Cadence of the future." Signed by the company president. I worked for them early and then they laid me off when I completed a project. Then they hired me again later and I completed a couple of projects that no one else could. Then they asked me to be a manager at a firm they were acquiring. I didn't want to but they kept upping the ante until I couldn't turn it down. Like I told them, I couldn't do it all by myself and it took six other engineers to do half of it and it couldn't get done on time. Then Cadence followed me around and acquired the companies I engineered for as they were closing and I was leaving. Finally Cadence became totally dislocated as an engineering firm and now has a 30,000 sq foot building 10 miles from where I live that does nothing but manage companies all over the place.

My memory may be damaged in some things but the way my mind works on engineering projects wasn't damaged at all. Google, Facebook and Microsoft have all called me but the first two are too far away as a commute and Microsoft is in Renton, WA, near Seattle and Damned if I would move to the wettest spot on Earth.

On Thursday, January 30, 2020 at 1:59:12 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

The single advantage of Di2 is that you never have to worry about "cable stretch". Tuesday in the middle of a damn rain shower the front derailleur cable pulled out and I had to do the remainder of the ride in the small ring. While stopped for coffee I repaired it but I was afraid it might break again and send the cable through the spokes or something and disable me so I waited until I was done with all of the hills before shifting into the big ring. It worked fine but it might not have. And this is with Campy Cables - the best you can buy. I just completely changed that cable this morning. It required me to strip the handlebar tape off and then I discovered that the outer was pulling apart at the lever. So I repaired everything. I had to go through 8 inner cables before I found one that was long enough. I don't know where I could have gotten cables that short. Anyway they are in the trashcan now.

On Thursday, 30 January 2020 17:40:19 UTC-5, Tom Kunich wrote:
On Thursday, January 30, 2020 at 1:59:12 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

The single advantage of Di2 is that you never have to worry about "cable stretch". Tuesday in the middle of a damn rain shower the front derailleur cable pulled out and I had to do the remainder of the ride in the small ring. While stopped for coffee I repaired it but I was afraid it might break again and send the cable through the spokes or something and disable me so I waited until I was done with all of the hills before shifting into the big ring. It worked fine but it might not have. And this is with Campy Cables - the best you can buy. I just completely changed that cable this morning. It required me to strip the handlebar tape off and then I discovered that the outer was pulling apart at the lever. So I repaired everything. I had to go through 8 inner cables before I found one that was long enough. I don't know where I could have gotten cables that short. Anyway they are in the trashcan now.

Gee whiz Tom. It's a simple matter to turn an adjusting screw to limit the front derailleur to the middle ring.

Was this on a regular derailleur or on your Di2 setup? Why couldn't you just reattach the cable to the front derailleur at the side of the road? That should take only a few seconds.

Cheers

On Thursday, January 30, 2020 at 2:40:19 PM UTC-8, Tom Kunich wrote:
On Thursday, January 30, 2020 at 1:59:12 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

The single advantage of Di2 is that you never have to worry about "cable stretch". Tuesday in the middle of a damn rain shower the front derailleur cable pulled out and I had to do the remainder of the ride in the small ring. While stopped for coffee I repaired it but I was afraid it might break again and send the cable through the spokes or something and disable me so I waited until I was done with all of the hills before shifting into the big ring. It worked fine but it might not have. And this is with Campy Cables - the best you can buy. I just completely changed that cable this morning. It required me to strip the handlebar tape off and then I discovered that the outer was pulling apart at the lever. So I repaired everything. I had to go through 8 inner cables before I found one that was long enough. I don't know where I could have gotten cables that short. Anyway they are in the trashcan now.

Did the cable slip or break? Fixing a slipped cable should take about ten seconds with a pocket tool, minus the time for slapping yourself for under-torquing it in the first place. The benefit of Di2 is lack of cable sticking under the BB or elsewhere, particularly for those of us who ride in the rain much of the time -- although my commuter swamp-mobile has cable shifting, and it works fine until the BB gets massively caked with mud, leaves, etc. My friend has chronic problems with cable sticking on his CX and rain bike, and he's fastidious with cleanliness and lubrication, so the problem varies by bike. Di2 is also nice for crisp shifts, but otherwise its a novelty.

-- Jay Beattie.

On Thursday, January 30, 2020 at 3:13:56 PM UTC-8, Sir Ridesalot wrote:
On Thursday, 30 January 2020 17:40:19 UTC-5, Tom Kunich wrote:
On Thursday, January 30, 2020 at 1:59:12 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

The single advantage of Di2 is that you never have to worry about "cable stretch". Tuesday in the middle of a damn rain shower the front derailleur cable pulled out and I had to do the remainder of the ride in the small ring. While stopped for coffee I repaired it but I was afraid it might break again and send the cable through the spokes or something and disable me so I waited until I was done with all of the hills before shifting into the big ring. It worked fine but it might not have. And this is with Campy Cables - the best you can buy. I just completely changed that cable this morning.. It required me to strip the handlebar tape off and then I discovered that the outer was pulling apart at the lever. So I repaired everything. I had to go through 8 inner cables before I found one that was long enough. I don't know where I could have gotten cables that short. Anyway they are in the trashcan now.

Gee whiz Tom. It's a simple matter to turn an adjusting screw to limit the front derailleur to the middle ring.

Was this on a regular derailleur or on your Di2 setup? Why couldn't you just reattach the cable to the front derailleur at the side of the road? That should take only a few seconds.

Cheers

Campy Record. And it was set up so that it worked fine on the small ring. Though it rattled slightly in the 11 tooth cog.

On Thursday, January 30, 2020 at 3:20:41 PM UTC-8, jbeattie wrote:
On Thursday, January 30, 2020 at 2:40:19 PM UTC-8, Tom Kunich wrote:
On Thursday, January 30, 2020 at 1:59:12 PM UTC-8, Andre Jute wrote:
On Thursday, January 30, 2020 at 3:38:01 PM UTC, Tom Kunich wrote:

The older Di2 doesn't download later firmware. It also doesn't have automatic shifting etc. In my opinion that is entirely unnecessary.

Again, given under the condition that my Di2 system is totally different, intended for commuters:

The Di2 "smoker" system gave you a choice sides the normal settings of sporting etc of full auto or assisted electronic shift where you would have to press a button to change gear. I found that when in the press-a-button mode I didn't take all that much less time over a circuit I rode every day of the week than when I had manual hub gears. The reason was, and is, that because I don't have the gift of cadence, I hang on to gears too long, just mashing through. However, in the full auto mode I knocked minutes off a 40m ride, a really appreciable difference since those days I didn't ride for fun but for necessary exercise and just wanted to get the hell back to my work.

Of course, if you have a good cadence and are used to changing gears for many small steps, the difference will not be as big as it was for me, but I'm loath to believe will be negligible.

Andre Jute
I started cycling too late to be a roadie

The single advantage of Di2 is that you never have to worry about "cable stretch". Tuesday in the middle of a damn rain shower the front derailleur cable pulled out and I had to do the remainder of the ride in the small ring. While stopped for coffee I repaired it but I was afraid it might break again and send the cable through the spokes or something and disable me so I waited until I was done with all of the hills before shifting into the big ring. It worked fine but it might not have. And this is with Campy Cables - the best you can buy. I just completely changed that cable this morning.. It required me to strip the handlebar tape off and then I discovered that the outer was pulling apart at the lever. So I repaired everything. I had to go through 8 inner cables before I found one that was long enough. I don't know where I could have gotten cables that short. Anyway they are in the trashcan now.

Did the cable slip or break? Fixing a slipped cable should take about ten seconds with a pocket tool, minus the time for slapping yourself for under-torquing it in the first place. The benefit of Di2 is lack of cable sticking under the BB or elsewhere, particularly for those of us who ride in the rain much of the time -- although my commuter swamp-mobile has cable shifting, and it works fine until the BB gets massively caked with mud, leaves, etc. My friend has chronic problems with cable sticking on his CX and rain bike, and he's fastidious with cleanliness and lubrication, so the problem varies by bike. Di2 is also nice for crisp shifts, but otherwise its a novelty.

-- Jay Beattie.

It looks like it cut one wire out of the bundle allowing it to reduce in diameter and slip out of the lock-screw. That is odd. When I was redoing everything this morning I discovered that the shift outer had pulled back out of the lever slightly. I think that it was putting excessive friction force on the inner which caused it. Even the wire tip fitting fell off. So perhaps it sawed it's way through the cable way up at the lever and pulled out at the far end. I'm using an integrated bar/stem that is nice but really puts the outer cables through a lot of direction changes. Almost straight up at the head end where the shift inner enters, almost a right angle turn at where it is shoved into the lever slot and then it weaves through the inside of the bar. This time I was extremely careful that it was all running a smoothly as it could.