Shimano cranks failing. Time to do something about it?

If the shop called Shimano, they would have sent you a replacement crankset for free even if it is out of warranty. We have done this for at least two customers, one of whom had this issue with a very old 6800-series crankset and Shimano sent him a R8000 replacement along with a new FD to match.

Imagine if instead of Hollowtech they’d called it Gluetech?


What’s the failure rate?

No one has died from cranks failing, so as noted it is a false analogy. I am not even sure what types of injuries have been reported. I have not personally heard of any, but that is just my knowledge base.

You also have twice now skipped over the points re: the failure rate you have claimed. What data are you basing your claim on?

Shimano made the choice to start gluing cranks together, instead of melt-forge or otherwise non-gluing technique. Why glue? It’s probably cheaper. Could they make a crank of the same weight without gluing the cranks together. Of course they could and they have.

Actually, consider this: we all want threaded BBs in our carbon bikes, right? You can’t thread carbon, so you have to put an aluminum sleeve in the frame. That’s attached by glue. Your rear triangle was almost certainly manufactured separately from the front triangle, and they bond those with glue. Alloy steerer into carbon fork probably also bonded with glue. Pedal inserts in carbon cranks also bonded with glue. They make glue in a bunch of different strengths ranging from Elmers all the way up to can attach an aluminum BB sleeve into a carbon frame. Just because the two halves are bonded together doesn’t mean that’s an inherently bad method. It could be they used the wrong type of glue, or it’s a manufacturing process issue.


The resin that holds the fibers together in your carbon fiber bike is, essentially, glue.


I injured myself by snapping a crank once: I was out of the saddle, climbing when the head of my right crank snapped off, pitching me head first onto the road.

My helmet smashed into pieces, I had minor concussion and abrasions. Without the helmet I’d have been in a sorry way. Fortunately this was the last hill before home.

FWIW the crank was an old Campagnolo Record with the 144 BCD, they were known for this failure. Since the crank was about thirty years old when it snapped I just thought “I should have checked that”.


This thread is gold… exploding engines killing whole families and cranks snapping throwing people forward where their helmet explodes into bits. :rofl:

Just to provide a little more support for the accusation, my beautiful Dura Ace FC-9000 cranks developed a hairline fracture on the spider where it meets the chainring and it migrated toward the center of the spider. I began noticing the sensation that my right foot was slightly swiveling around the pedal stroke. I took my bike in to the shop, and one mechanic verified it with my own pedal and then another pedal in the shop. But they refused to try to warranty it. After buying a new set of FC-9100s, I started cleaning up the old pair to gift them to a friend to try out. That’s when I discovered the fracture. I wrote a long-winded explanation of the situation and sent it and the cranks to Shimano on my own dime. The company warranty department turned it around and sent me (a second new) pair of FC-9100 cranks in record time. I was irritated, thankful, and perplexed all at once. When I tried contacting Shimano North America again to get some sort of explanation of the failure, my request was met with silence. Luckily, someone was interested in a new pair of FC-9100 cranks and I was able to unload them on eBay for a small loss. After almost two years on the new cranks, I haven’t had a single problem with them. I can’t be sure, but I wouldn’t be surprised if they solved whatever the problem was in the release of the Dura Ace 9100 groupset. And that may have happened with the Ultegra, too.

1 Like

To be clear, I am not saying that injuries are not possible (likely?) to occur with a crank failure….it isn’t hard to see how they could happen.

I was simply saying that I personally had not heard of any related to this issue w/ Shimano cranks….even from reports I have seen on various forums.

I didn’t think you were but I thought an example of just what can happen might be interesting.

Mine was admittedly a bit of a freak accident: One moment I was cranking up a short but very steep hill in a fairly big gear*, the next thing I knew I woke up lying on the road with my pedal and part of the crank still attached to my foot.

I was leaning a long way forward as I typically did on that last hill, I think I must have hit the handlebars on the way down and pitched forward from there.

  • The crank would have been the standard 42 / 52 and the freewheel probably 13 - 23. That’s what we rode back then: only touring bikes had low gears.

Sure, Campagnolo carbon cranks are also held together by a resin one could call “glue”. My point is that for cranks specifically, is there anything wrong with previously made Shimano cranks which were also hollow and not glued or bonded together? Is there a functional purpose or true measurable advantage to the clamshell design, the hollow spider designs? If there isn’t, then Shimano has chosen the design simply to make their parts proprietary or even for simply being unique for marketing purposes. I’m not even saying there is anything wrong with making parts designed for aesthetics or for making something proprietary.
Personally, FC-7800’s are my favorite design.

If a crank failure could produce a movement anything like chains skipping over worn chainrings, then the abrupt weight transfer has a very high probability of causing a rider to lose control. This exact thing happened to me and I broke my hand.

FC-7800 cranks featured a welded cover and its spider’s five arms were not hollow. So maybe their last “safe” design.

Those were individually checked by X-ray inspection to make sure the welds were o.K.

1 Like

A very high probability based upon your single anecdotal example? That’s not how probabilities work.

Hey, if we’re doing anecdotes, the last crank I broke was a SRAM X01 crank when I cased the second big double at the bottom of Crabapple Hits in Whistler. I bruised my pride and rode it out on the left crank, no harm done.

Shit breaks, life goes on… worry about your bike handling skills more than Shimano’s manufacturing choices.

Well, tbf, that the mechanism of failure he’s describing has a high probability of causing an accident is beyond reasonable dispute, I’d say.

However, that said mechanism is actually that common or likely is much less clear.

Basic risk assessments look at both severity and probability: here, the first is variable and the latter is unknown, but it doesn’t seem terribly probable.

When you boil it all down, essential point here is that when travelling along a paved road, with traffic, at speed, any sudden failure is potentially highly dangerous, be that a saddle rail, a tyre sidewall, a seatpost… you get the idea.

Beyond regular inspection of your bike, and not buying equipment you don’t trust, there’s not a lot you can do.

Personally, I worry far more about inattentive and aggressive drivers and the new potholes that seem to be appearing everywhere at the moment.

1 Like

You could say that, but you’d be just as guilty of making faulty assumptions as he is.

In reality, if you’re in a pack sprinting away from the start line then sure that would lead to a pretty spectacular crash… like what happened at Waaslandcross-Sint Niklaus earlier this year. But, in practice, if losing a pedal or a crankarm causes you to completely lose control in any situation other than a full sprint then you really need to work on your bike handling skills.

You have assumed I’ve said completely lose control - I didn’t. I said cause an accident. That can be caused by a 12 inch deviation off line in a pace line. That can be caused by failing to see the stop light or pothole as you’re distracted. That suddenly and unexpectedly losing a pedal or piece of crank has a good chance of causing an accident is not a faulty assumption, it is a reasonable one.

As I noted, the chances at said accident turns out to be serious are much harder to define. To continue the examples, you might nudge shoulders with someone and you both wobble and ride on, or you might take down the whole pack, Opi Omi style. And, as I also noted, the chances of it happening in the first place seem pretty small.

But simply to argue that a crank arm failure per se isn’t dangerous is a deliberately perverse argumentative stance.


Accidents are caused by losing control… we can play pedantic games over what constitutes complete loss of control vs loss of control to sufficient to cause an accident but, in either case, if you’re taking down the pack anywhere outside of a sprint situation because you broke a crankarm/slipped a pedal/skipped a cog or anything else that causes a sudden transfer of weight from both feet to one foot, then your bike handling skills are the real issue and you have no business being in a pack.

It’s legitimately funny how big a deal y’all are making of this… I remember when everyone was tossing carbon SRAM cranks on their DH bikes and not a weekend would pass without someone breaking theirs and it rarely resulted in someone hitting the deck, and almost never resulted in anyone crying about it… let alone a “time to do something about it” forum.

Broke a crank arm = skipped a cog. Lol.

1 Like