Why do Shimano (Ultegra & Dura-Ace hollowtech) cranks keep failing? Analysis.

Peak Tourque hypothesizes it’s moisture ingress leading to corrosion and disintegration of the glue that binds the crank arms together, made worse for people living near the equator due to heat/cold cycles causing condensation. For Shimano crank gore see: https://www.instagram.com/thanksshimano

But why don’t 105 R5800 or R7000 seem to fail like this? Different glue used? More material and stronger? Different design preventing moisture ingress better?

Don’t get me wrong, I love Shimano parts (use R5800 cranks myself) and I hope posts and videos like this help make sure they fix the problem in the future and help protect current owners from accidents so you know what to look for inspecting your cranks and even how to help ensure they don’t corrode – if that is the reason.

More material and stronger?

Is it not a different process used?
Like the difference between M9000 & M9020 crank arms?

Could very well be a different design / process, not an expert on this.

Absolutely didn’t know this was a thing. Thanks!! Something else for me to worry about 😧 ( 6yr old ultegra crank owner)

Why do Shimano (Ultegra & Dura-Ace hollowtech) cranks keep failing?

Because there are tens, or hundreds of thousands of them in use every day?

One anecdote doesn’t make a statistic.

Because there are tens, or hundreds of thousands of them in use every day?

This. Do we even know that they keep failing, beyond a fraction of a percent?

It’s certainly reported more than any other current crank design failure that I’ve read about (still not suggesting it’s a disproportionately high number fail though).

Wasn’t even aware of this issue, we’ve sold hundreds of bikes equipped with Ultegra chainsets, and had no returns on any of them. Strange.

What cranks are you changing them for ?

Am I the only one did in videos like this seem way longer than they need to be?

CBA watching 20m.

This. Do we even know that they keep failing, beyond a fraction of a percent?

I think so. We don’t see 105 cranks failing and we know they have a different construction where the tube is pinched shut at the pedal eye whereas R6800 and R9000 are bonded, and there are dozens of threads and anecdotes about this. Shimano even acknowledges the problem! https://www.bikeradar.com/features/shimano-crank-failure/

” dunno but here in Thailand there seems to be an increasing number of failures, this happened to one of our club members a few weeks back, Shimano swapped it out under warranty. Failed after just 2000km/5 months.”

“It’s a raging failure. I’m in the US and have seen this on several peoples Ultegra cranks. Shimano needs to do a recall or get badly sued when someone gets hurt.”

https://weightweenies.starbike.com/forum/viewtopic.php?t=143022&start=30

Wasn’t even aware of this issue, we’ve sold hundreds of bikes equipped with Ultegra chainsets, and had no returns on any of them. Strange.

Maybe not so strange. If like PT is hypothesizing it’s only happening in regions where condensation is likely to build up in the crank arms (like Thailand or parts of the US). It could however also be bad batches from Shimano. I haven’t seen any examples of broken Ultegra or DA cranks where there haven’t been visible corrosion.

What cranks are you changing them for ?

Not changing them, R5800s don’t break like this.

Shimano even acknowledges the problem!

Read it, cant find anything to suggest that.

“It’s a raging failure. I’m in the US and have seen this on several peoples Ultegra cranks. Shimano needs to do a recall or get badly sued when someone gets hurt.”

Remember than when people say this on the internet they actually mean “have seen this on several peoples cranks on forums in a country of 300million people”. They’re not likley to be talking about failures within their circle of friends.

The post under it….

Having worked in a large, road oriented bike shop through the entire 6700, 6800 and now 8000 eras of ultegra I have never seen an Ultegra crank failure, or even any incipient signs of this kind of failure.

I’ve had a 6800 crank fail 2 years ago, and a 8000 crank fail over the winter.

Both cranks warranted by shimano/ madison via my lbs.

Shimano even acknowledges the problem!

Ok they didn’t, misread. Not surprising, I guess no problem is acknowledged short of a recall (which they did in 97).

Remember than when people say this on the internet they actually mean “have seen this on several peoples cranks on forums in a country of 300million people”

You don’t know this. You can assume it’s the case, or you can assume it’s not which is a smarter choice – better safe than sorry.

Having worked in a large, road oriented bike shop through the entire 6700, 6800 and now 8000 eras of ultegra I have never seen an Ultegra crank failure, or even any incipient signs of this kind of failure.

Again, this can be explained by regional climate differences andor batch problems.

So, if you own a Ultegra or DA crank, you can grease the cracks to prevent moisture ingress, store it inside or avoid storing it outside if there are large temp jumps between night and day etc.

I’ve had a R6800 go – front and back delaminated. Just over two years use. Out of warranty and had to buy a fresh one. Put an R8000 on – 14 months later – propagating crack below the pedal spindle followed by delamination below the crack line. The pedal felt ‘wonky’ for about 50km I thought it was the cleats / shoes. Whole thing failed when I was about 30km from home, fortunately a call out for a Thunderbird got me home. Quite lucky it went when it did, only just back from a bike trip in Norway a week or two before. It would have messed with the logistics a bit.

There didn’t seem to be any signs of corrosion – in the same way as you might see white powder in ski/ walking poles if you don’t strip them to dry. Each time it was only 8000-10,000km of use. I know of one other person who has had the same issue.

The photo below is after I kept riding it to get to shelter / phone signal.

There didn’t seem to be any signs of corrosion

Inside or outside?

Managed to get a photo to show – the polishing is more to do with rub – this is the second one. The first one I’d have down as something to do with the bonding process. The second – I don’t know. It would have to have cracked within the crank itself to then go? The front plate didn’t seem to have any signs of damage. I’ll qualify that with I’m a geographer not a metallurgist! Could be my weight, rough roads and a love of the local pave hasn’t helped. The crank in the photo went back to the supplier. The first one went to recycling.

The crack opened out to full width in 25km. Prior to that, as before it just felt wonky but I didn’t think to look at the back of the crank. The first started with a really strong metal creak at one point in the pedal stroke. I thought it was the BB.

Bike is washed after each ride but then stored in the kitchen which is dry / dehumidified regularly.

Another of same

https://handsonbike.blogspot.com/2016/10/shimano-ultegra-6800-vs-shimano-105.html

Think the bonding saves 15 grams or so: penny wise, gram foolish :p

Someone In the comments

“Thank you for this! I just destroyed my Ultegra 6800 cranks after 4,041 miles and 27 months. The epoxy came loose on the crank arm and are no longer usable. Having 105 on two other bikes, I have decided to replace the Ultegra cranks with 105 5800. The weight difference is negligible and the cost of replacement chainrings is cheaper.”

This went round the club’s WhatsApp around the same time mine failed – think it’s originally from a bike owners Facebook group.

Shimano’s hollow cranks have been doing this as long as they’ve been around- I remember XT square tapers doing it back in the early 2000s. But, as now, it’s an exceptionally rare event and there are so many of these cranks about that it’s not actually a very big problem.

Meh – I bought some 105 recently instead of Ultegra but that was only because it’s available in silver.

0.5-1% failure rate might be realistic.

“cat3memes on instagram posted a poll on this topic yesterday, asking how many people have had one of their shimano cranks fail. Today the results were posted… 223 out of 1985 responses. That’s 11%.

Let’s be very conservative and say that everyone of his 34,000 followers has had 2 of the shimano cranks that are prone to the delimantion issue. That would be a .3% failure rate. I don’t know how many of these cranks are out in the market, but it’s a lot. And that’s the conservative view. The more realistic view is maybe .75 cranks per follower, almost 1%.

Think Mavic R-Sys wheels, some guy experienced a failure (that was probably due to some kind of crash damage) and the whole concept died. They released another version, I never heard of another failure, but the concept still died. Zipp recalled a entire hub at a very high cost to them after less than handful of them experienced a catastrophic failure.

These are catastrophic failures and where there is smoke, there is always fire. Why is this OK? I would much prefer a company like Zipp that is transparent and announces a recall vs Shimano that sweeps everything under the rug to present a perfect image of consistency and high quality.” https://weightweenies.starbike.com/forum/viewtopic.php?f=3&t=143022&start=135

And to get the terminology right: It’s only Dura-Ace and Ultegra that are affected since they’re hollow-bonded. 105 cranks and below are hollow-forged so they’re fine.

Happened to me too, last summer but because mine were 3 years old Madison wouldn’t warranty them. Mine delaminated like the ones in the picture from oldagepredator. It also took me a while to work out what was wrong – I thought it was wonky pedals and bought new ones before I found the cracks in the crank

That would be a .3% failure rate.

On the face of it, but that’s not a test of reliability. Engineers will decide on how long a product will should last, and who many of them should make it to that point, then build in a measure of safety margin.

Say you expect a crank to last I dunno, 30,000km and you expect 95% to succeed, that’s a reliability test, not whether some bloke on the instanet can get people to tell him about their crank failing.

I’ve had ultegra crank fail the same way.