1

The bicycle component manufacturer Shimano makes crankarms named "Hollowtech".

Hollowtech crankarms are meant to be both strong and light. As the diagram below (source: Shimano) shows, the outer shell is made from an aluminum alloy and the inside is hollow.

hollowtech

On closer look it turns out that the cross-section of the crankarm is not made from a box section, but from two C-sections. (Source: bikeradar)

two c-sections

Another (older) style of Hollowtech consisted of a C-section closed off by a plate.

failure mode of hollowtech

The two parts are glued using epoxy. It's understood that epoxy, once it sets, is more than adequate for the task of carrying the weight of a cyclist on the crankarm, but even if epoxy does bind well with a (perhaps roughened) alloy, wouldn't the difference in the stress-strain curves of the two materials mean that when the alloy bends, it is the epoxy that will take all of the load? Since it's used here only as a binding agent, it would appear this is a recipe for failure. Do you understand how epoxy can reliably bind two parts of a crank arm, and can you explain it?

Sam
  • 385
  • 2
  • 7
  • 1
    A hollow crank should be worse for aerodynamics than a solid one with the same strength (but higher weight). I don’t think it’s a hoax or marketing trick. Hollow crankarms (even if they are made from two or more parts) are a valid way to reduce weight while achieving the same strength. – Michael Nov 08 '22 at 15:25
  • Did Shimano ever mention aerodynamics? – EarlGrey Nov 08 '22 at 16:36
  • @Michael I'm not sure how you reached this conclusion. Your intuition is clearly quite different than mine. But perhaps once again this is a question more suited to an engineering forum than a cycling one. – Sam Nov 08 '22 at 17:49
  • 4
    This question seems to be fishing for supporting answers IMO. In my mind forces that would be required to damage hollow cranks are abnormal and would really have to be the result of impact, bad technique or plan miss-treatment, for that alone I would be more than happy to wreak a crank set then potentially ripping the BB out of the frame (being dramatic of course) –  Nov 09 '22 at 03:45
  • 2
    "no one in construction manages to reach the torsional rigidity of a closed section" - 1) yes they do, and 2) most of the time they don't because they don't need to. For a building an I-Beam and a Box beam are both just as rigid. – whatsisname Nov 09 '22 at 03:57
  • 2
    I'd be very careful with calling a product a "hoax". You may find yourself at the wrong end of a costly lawsuit. – cmaster - reinstate monica Nov 09 '22 at 08:15
  • 1
    Btw, the technology to build truly hollow cranks is there: Simply weld the two pieces together. That Shimano didn't do that makes one thing very clear: They didn't really care about actually strengthening the piece when they decided to close off the cavity. – cmaster - reinstate monica Nov 09 '22 at 08:22
  • 2
    @cmaster - reinstate monica: or because epoxy or whatever else they used gave a better bang for the buck. Welding would sacrifices a lot of the strength from the forging process. – whatsisname Nov 09 '22 at 08:31
  • @whatsisname True, you lose all the work hardening from the regions that get too hot during welding. Nevertheless, especially with the non-vertical contact surfaces shown in the last image, any epoxy, or other glue solution, cannot hope to compete with the strength of a weld. At least, not when shearing forces are involved, which is exactly the failure mode shown in the first image. At the very least, you need a brazed connection to withstand significant shearing forces, but I have never heard of a process to braze aluminum. – cmaster - reinstate monica Nov 09 '22 at 08:52
  • When the Hollowtech cranks first appeared I understood Shimano would have used some kind of forge welding process (might been misunderstanding or false advertising). Glue in location this critical feels odd. –  Nov 09 '22 at 13:29
  • Looks like I was right: the older design doesn't have obvious glue seams and fails in a different way: http://pardo.net/bike/pic/fail-001/FAIL-170.html –  Nov 09 '22 at 13:42
  • I used to look at bikes built with SRAM drivetrain but with another brand's brakes, and wonder "oh, they cheapened out"—until I learned of the advantages (https://bicycles.stackexchange.com/q/86517/48599). I also used to look at bikes built with Shimano components but with non-Shimano cranksets and regret the ease with which replacement components can be found with specific part numbers (even if at price); I used to marvel at the elegance of the spider and bemoan the incomplete. Yet I now see, as this question illustrates, that I would really prefer a sturdier crankset. – Sam Nov 09 '22 at 14:47
  • @cmaster-reinstatemonica My Properties of Materials course never discussed the use of epoxy resins as a structural element in dynamic applications—I'm not sure whether that was an oversight or an incredible application. Epoxy is of course widely accepted against concrete, and binds exceptionally well with porous surfaces. At this point I am mainly concerned about how Shimano could have possibly used epoxy (anyone care to provide a reference that they did use it to bind two-part cranks?), and how they bet that they could manage to avoid costly (personal or wider) lawsuits. – Sam Nov 09 '22 at 14:52
  • @cmaster-reinstatemonica: it doesn't have to "compete", it doesn't need maximal strength, it just needs good enough strength. Here's a different manufacturer that also uses an adhesive: https://www.youtube.com/watch?v=UHU3EaAPUR4 And if you think glue isn't good enough, just remember we have entire bike frames that are made of carbon strings, and glue. – whatsisname Nov 10 '22 at 03:16
  • @whatsisname True, but we don't step on the top bar and push with all our might. Doing so would be "risky". A crank arm, on the other hand, needs to support someone stomping on a climb while muscling the handlebar. – Sam Nov 10 '22 at 03:35
  • @Sam: you probably could, and anyways if that was intended, it would be designed so. They make fighter jets and race cars structural elements with the stuff. These crank arms are designed by people that work on them all day every day, they know more about it than you. – whatsisname Nov 10 '22 at 06:47
  • 1
    @whatsisname Carbon fibers have a surface composed of C-H groups. A few impurities in this chain of C-H groups suffices to produce actual atomic bonds from the resin to the fiber, or cross-link the fibers into a net that can bond mechanically to the resin. Not so with a metal surface. Chemically, that's a sea of free electrons with some positive ions dispersed in it. That's really hard to bond to in a shear resistant way. Your best hope is to bond mechanically to a roughened metal surface. And again, the first photo shows that whatever bonded the parts, it failed. – cmaster - reinstate monica Nov 10 '22 at 07:51
  • "Once the parts are treated and machined, a fully automated system of computer controlled robots applies the glue and bonds together the crank arms.... This process lets them use longer, larger hollow sections to save more weight than they could with a one-piece hollow forging. ... Once the parts are pressed together in a jig, they go into the oven. Shimano won’t disclose baking time or temp. The bonding agent is their own formula they worked with the glue manufacturer to develop." https://bikerumor.com/shimano-factory-tour-this-is-why-theyre-masters-of-the-metal/ – Vladimir F Героям слава Nov 10 '22 at 11:09
  • As an aside, anyone know how the Sora R3000 crankset is built? –  Nov 10 '22 at 12:18
  • @newroadie I do not think it is hollow. Just a forged single piece of metal. – Vladimir F Героям слава Nov 10 '22 at 15:24

2 Answers2

8

It does nothing for aerodynamics, the cavity makes the thing bigger and hence producing more drag (a very hypothetical reduction of Cd cannot change that). The point is simply the weight.

Perhaps it is also somewhat better to talk about stiffness rather than just strength (which may be an ambiguous term). You need the cranks to be as stiff as possible when doing sprints on high-end cranks.

Has the manufacturing process for R8000 series improved over R6800? Who knows, but the combination of two parts is still there and the failure is still possible. For example, https://yacf.co.uk/forum/index.php?topic=112701.0

Vladimir F Героям слава
  • 181
  • 3
2

Agree with Vladimir. Note that the join between the two C-section halves can be created with a deliberate interference (force) fit involving the roughened surfaces in such a way that when they are driven together in a press, the pieces gall and seize. This means the parts are in a sense friction-welded together and thus the epoxy does not have to carry all the imposed loads.

Another thing to consider is that if the interference fit creates residual compressive stresses in the assembly, then all you have to do to prevent the stress on the glue line from ever becoming tensile (which is how it would fail in service) is to ensure that the magnitude of the residual compressive stress exceeds the maximum tensile service stress.

niels nielsen
  • 13,033
  • 1
  • 11
  • 30
  • Intriguing. You're somewhat upturning my thinking about loads in cycling. Normally the bike is the fluffy thing (~10kg) and the cyclist (being many times that) is relatively a monster. You're effectively saying this is a solution because in the grand scheme of things even the limit 110-kg cyclist is a fluffy thing compared to what a box-alloy crank can potentially handle. – Sam Nov 19 '22 at 15:14
  • But to be clear: there is one nagging point you do not address. Two geometrically identical beams, one made from the alloy and the other from epoxy, would have vastly different deflections under the same load. That's to be expected, for one is a metal, while the other is closer to glass. By the time a beam made from BOTH materials deflects, the alloy would be taking NONE of the load. The only remaining explanation is that the epoxy is merely a micro-scale glue. It basically shatters everywhere and only the alloy is load-bearing. Is that also your understanding? – Sam Nov 19 '22 at 15:14
  • @sam, there are many stressed airplane components held together with glue. There are bicycle frames that are glued together too. I do not know enough about how this crank arm was designed to comment further on it- NN – niels nielsen Nov 20 '22 at 01:28
  • Fair enough... Moving on... – Sam Nov 20 '22 at 03:39