Shimano HollowTech crank arm failures
 

Here is an updated analysis of crank arm failures. I have yet too see this on my Ultegra 6800 crankset yet, but keeping doing regular inspections.

With influencers needing money, I'd wouldn't be surprised if this just isn't BS to get youtube hits. Not that anything is a lie, but mostly about a something that occasionally happens, but statistically is a non issue. Solid cranks of all sorts have broken too.

I'll wait to see what comes up from the experts that don't depend on youtube for income.

Haven't seen the video (will comment first, to see how stupid I am :) ) - but I've seen those cranks.

When the pedal is at 6 o'clock, and you weigh it, the crank arm needs to resist bending towards the inside of the bike (closer to it).
The shape of those cranks is very narrow along that axis - while they are very wide along the axis that doesn't get all that sressed.
A shape closer to square cross-section would probably be better suited for such loads.
These cranks are designed to look cool, aero. And they are bonded with glue (hollow on the inside, and glued, not welded, as far as I know).

Aluminium gets fatigue cracks if it's not "overengineered" enough to prevent much bending (elastic deformation, not permanent).
Glued aluminium should also allow for some micro-openings that let moisture in. If there's combination of steel and aluminium, that's a beautiful place for galvanic corrosion to start.

The shape alone is enough to have me raise an eyebrow but think: they may have tested it, it may still be strong and stiff enough along the critical load axis - and they risked with extra width (and weight) just for show (more aero, more cool?).

But if they break a lot, it's probably not good enough.
If they break more often in the climates with more moisture/rain - there probably is some combination of aluminium and steel in those cranks, and galvanic corrosion does its job.

When I see failed parts first hand, I usually inspect them and try to figure out why they failed. I haven't seen these broken yet, but they aren't very widely used in my city.

You know that more and more aluminum airplanes that you ride in are glued together don't you? I was flying an airline transport certificated plane back in the 90's that most of the aluminum skin panels on the wings were glued. The very same airplanes I flew then are still in service.

So while you made basically true statements, they still don't tell us if the Shimano cranks are having any more failures than any other type of crank sold today.

Hmmm,

I stopped with Hollowtech and Octalink, which does have some issues, but very different from above. I think the Octalink versions were welded.

I see some notes that heavy anodizing will prevent some of the galvanic corrosion, although I'm not sure about anodizing under the aluminum/steel interface. Perhaps they could anodize, then heat and shrink fit.

It may be that my Octalink hollowtech cranks can be welded as they become a sealed capsule. Instead, the 24mm cranksets require a steel or titanium axle to be permanently bonded to the crank (thus inserted from outside-in, then laminated over).

The torsion along the glue joint appears to be problematic.

It is possible that the Ultegra cranksets have a little extra redundancy that the Dura Ace cranks don't have. I'll try to see if I can figure out how the older Ultegra is bonded.

It's disappointing to see Shimano trying to weasel out of this issue. They are ruining their Dura Ace brand.

Any of you guys know someone who has broken a crank arm JRA?

One of my personal friends (not just Some Guy online) broke an Ultegra crank. I was shocked as I didn't expect this type of failure from Shimano. It's just one data point, but there it is.

He's average height/weight/power. This was just on a road bike.

https://cimg8.ibsrv.net/gimg/bikefor...83eff56789.jpg

No idea about aeroplanes.
But when it comes to bicycle cranks, I suppose that profile Shimano chose, combined with the materials used and the chosen method (glueing) could cause problems.

I agree. Without aggregate data (in thousands, preferably tens of thousands), we can't conclude the failure rate.

Having said that - most cranks are poorly designed. Starting with the pedal mounting interface which seems to be one of the few things in the cycling industry that is standard and uniform across different manufacturers and over decades. Too bad it's a poor design (that's probably considered to be "good enough if you aren't heavy and don't ride really hard"). :)

I would strongly suggest watching his video before commenting. Sachin (Hambini) is an aerospace engineer, specializing in aerodynamics, who works for (I think) Airbus. So, no, he doesn't need the measly YouTube income.

He only started his YouTube channel when a couple of his high-end carbon frame bikes started failing. Typical engineer - he researched and was astonished at how many carbon frames are made to such poor tolerances for press-fit bottom bracket bearings. Worse still, after he pointed this out for a certain brand, that brand retroactively INCREASED the tolerance limits so they could then claim it was not their fault, but rather the LBS or misuse by the owner. Ugh. (He also polled a number of bearing manufacturers, who all stated they will not warrant their bearings if used in such sloppy tolerances)!

I'm also a mechanical engineer, and can vouch for his analysis.

I would fully expect ShimaNO to create a 'new style' crank and then run away and duck out from the failed Hollowtech units

Dyna drive pedals = Positron shifters = front freewheel system = pitch 10 = self centering Tourney brakes = AX 'aero' parts ( the waterbottle accounted for the biggest percentage of the 'aero' b enefit ) = ALL of these ( and many more ) were abandoned by ShimaNO and left countless cyclists 'holding the bag'

I'm a little surprised that Dura Ace hasn't gone to Carbon Fiber with Titanium inserts.

But, of course, that is also a learning curve.

For all the engineering props that the media likes to place on Shimano the company is a very marketing driven one. I would suspect that the marketing side has the upper hand when decisions are being made about spending millions to bring a new design to a profitable level. Shimano isn't the first bike company to suffer with bonded design failures. Andy

Here are a couple other videos.



Ok, the first video perhaps a little less professional than the OP's video. However, I have to wonder about the theory from the first post of galvanic corrosion occurring almost uniformly around the crank, rather than primarily around the crank spindle.

Here, the presenter suggests perhaps heat/cooling cycles and hotter parts of the world. Salt and coastal areas, and perhaps a few additional faults. FLORIDA MAN!!!

The last video has some more analysis.

In the middle of the last video, 6:40 to 8 something there is a discussion about 105 vs Ultegra/Dura Ace.

The 105 is "hollow forged" which I believe is the process from the Octalink cranks.

The Ultegra/Dura Ace is "hollow bonded" which is the new process that is failing.

So, if you are racing, perhaps pick up the Dura Ace, and replace at the end of the season. But, if you want to have a bike that lasts for decades, then get the 105.

One thing allowed by the Internet is that the entire customer base for everything is online in some form. Basically any failure rate here is not acceptable and should be rare and mostly accidental. But a steady failure mode across several generations of cranks out in the public eye is a real loser. They have a real engineering department, it’s not like they need Hambini to tell them how they got it wrong. Someone in there is screaming at the marketing and warranty people if they’ve got any self respect. It’s amazing no one has died and there hasn’t been a recall

So you’re saying is that a product should be made once, never improved and never refined?

You’re basically advocating for stone tools here, anything else is going to leave those poor cavemen ‘holding the bag’

Were there hollow crankarms in the 5-arm design too?
 

Does anyone know if the previous 5-arm generation was also hollow, as in crankarms? Not the spindle - it's somewhat misleading to believe it's the bonding with spindle that is the culprit here, HTII has been around for a while before news of these started coming out.

I came across Hambini longer ago for the BBs and I am afraid this is also the same marketing vs engineering take on the same. He makes a lot of sense in lots of his videos if one can abstract themselves from him sounding borderline infantile. Most people would prefer a well-spoken person with non-engineering degree sell them something shiny to listening Sachin's style presentations. Whether that's right or wrong, it's a matter of fact.

What I am saying is that ShimaNO produces stuff with big fanfare, flag waving and trumpets blaring ....then, when problems arise that show the NEW, IMPROVED 'original design' to be problematic they rather quickly change direction and abandon the product they previously ballyhooed

Tens of thousands ( if not 100,000's ) of bicycles with Positron PPS shift system were produced ....when Positron proved to be crap, ShimaNO utterly failed to provide back up support in the form of simple replacement cable assemblies ....tell ya what, give Shimano a call and ask them to send you a brown colored PPS cable assembly = crickets chirping = ( quick, bury your dead ) ....consequently most all of those Positron equipped bicycles are now taking up space in landfills, not to mention the number of potential cyclists who abandoned the sport due to the bad experience they had when using bicycles with these severely flawed parts that they were originally told were a really great improvement = NOT !

Have you ever actually worked on a bicycle with Positron ? ....Bueller ?....Bueller ?

Exactly Andrew R Stewart , a mega company like Shimano will have massive weight put on marketing decisions as speed of innovation and the corresponding market share are their life blood. Their Board carries the weight of making the trade offs, acceptable failure rate vs. the zero failure $$$ crank that Sram and Campagnolo will love as they scoop up market share.

My first work on Positron (or Negatron as we called them) was back in 1973 when the unit used a push/pull cable loop. Interesting that you mention Shimano stopping production and moving on from poor initial attempts and use the Positron as the example when it went through 3 versions that I know of over a 10+ year period. Not exactly dropping the goal IMO. Also, Shimano uses a trickle down of features as they change out a group/series (usually market driven reasons to limit a product to a few seasons only). Again not dropping the goals IMO.

I totally understand and have spent much time explaining to customers that their many year old item is no longer supported (and this is not unique to Shimano at all. Remember when front suspension was introduced and the manufactures {both US and foreign owned} would only support their forks for 3 years). I was brought up in shops that stocked der cages, anchor bolts and return springs and these days a repair is more often a replacement of a complete assembly. But I also have learned to choose what I get all riled up over and what is worth my negative energy.

Regardless of what you think of Shimano as a company their approach has worked so well they are that 500lb gorilla which now dictates the frame makers how to make that stuff. Andy

Hollowtech has been around for quite some time. The Octalink (Ultegra 6500 and related) were called "Hollowtech II".

Based on the current notes, there seems to be a leap from "Hollow Forged" to "Hollow Bonded".

The older Octalink cranks were the Hollow Forged technology. And the ones that are failing are Hollow Bonded. The 105 (4 arm) and lower models are the Hollow Forged (or potentially solid in the lowest).

https://cimg8.ibsrv.net/gimg/bikefor...fae691a31a.jpg

https://cimg2.ibsrv.net/gimg/bikefor...db90abb39a.jpg

https://cimg4.ibsrv.net/gimg/bikefor...131abe0201.jpg

For weights of "new" sets (2018), I'm seeing:
https://ccache.cc/blogs/newsroom/201...ght-comparison

Dura Ace R9100, Cranks, 614g
Ultegra R8000, Cranks, 674g
105 R7000, Cranks, 713g
There isn't much weight savings with the Campgnolo carbon fiber cranksets.

So the 105 crankset is about 100 grams heavier than the Dura Ace, with the Ultegra somewhere in the middle.

The bonded cranks appear to have a cavernous internal space. And I can understand why Shimano chose that technology. However, the forged cranks seem to have more aluminum in the areas where more strength is needed.

Ultimately, if I move to the Shimano 4 bolt cranks, it may well be with the 105 which apparently is somewhat more robust (although it may depend a bit on how the spindle is formed and inserted).

Unfortunately I'm not seeing a list of which cranks had the bonded and which had the forged technology.

I have an Ultegra 10-speed, 6700 crankset, and for the life of me, I can't find any bonding marks on the crank arms or spider. So, I believe that has the previous technology of hollowtech forged. :thumb:

The 6700 chainring, however, does appear to be bonded. However, the complaints I'm seeing are complaints about crank failures, and not chainring failures. In fact, the reviews of the hollow chainrings seem to be good, noting good weight to strength.

I don't have the Dura Ace 7800 crankset, but I think it may also be forged hollowtech (contemporary to the 6700, or slightly earlier). And, the chainrings appear to also be machined and not bonded.

There is a fair amount of frustration with companies dropping older models of technology, and not fully supporting the old parts.

Of course, that occurs in all industries. Oddly one can't still buy a 1920's Ford Model T. And, even looking at a model that has a long production run, a 2020's Corvette is very different from a 1950's Corvette.

What the auto industry has is a number of direct replacement aftermarket parts. Especially since things like starters, alternators, etc, were generally bought from third parties anyway.

We do see a lot of trickle down technology in the bike industry. And, some wear items like cassettes and bottom brackets are supported for quite some time.

And, these issues aren't just Shimano. Campagnolo has upgraded their "Record" line many times, and has left more than a few orphans behind.

SRAM has grown through buying up smaller companies. And, then they will periodically shed any technology that they don't like, and only keep what they like. They bought SACHS internal gear hubs, which they sold for a while under the SRAM brand, then dumped them all, including ones that had a unique niche.

It is hard to say how Shimano will deal with this crank failure problem. They may just try to ignore it, but it is very bad for PR, and ultimately will lead to some injury lawsuits. If I was them, I'd try to merge the Ultegra line with the forged 105 line (even at a slight weight penalty). Then work on both improving the engineering of the bonded Dura Ace line (less flex), as well as improving the gluing procedure and bonding agents. Possibly doing assembly under low pressure Nitrogen.

Have you been freeze dried, or doing hard time? Positron hasn't been around since like 1983.

Same here. Ultegra crank broke last summer after one season (a few thousand miles).