Hacker News Reader

Comment by top_sigrid | Hacker News Reader

While this is amazing, I keep wondering, why - even though in the 1940s - they didn't come up with the idea of a chain tensioner to compensate for different sprocket sizes. Instead they used this wheel suspension quick release mechanism which needed to be operated while driving, which just seems insane.

Weren't chain propelled systems a thing in other machines back then and a chain tensioner a known solution? Or were they just not suitable for a bike back then for some reason? Does anybody know something about this?

View on HN

It's because there was no freewheel. A chain tensioner only works well when there is driving force applied in one direction, not the other. Put any force backward into the pedals (e.g. to slow down) and the tensioner will be forced to its most slack position (or broken off).

Ok, this makes perfectly sense and is actually quite easy to understand. Thanks from a mechanical noob.

The article describes it as four sprockets on a freehub. One would hope that cyclingtips.com would get that fairly material detail correct.

It also mentions the rider having to pedal backwards to shift the chain among the sprockets. That would not be possible on a fixed hub without slowing to a stop and reversing. One suspects that would rather negate any benefits of this system over a flip-flop wheel.

The freehub was not invented for another 30 years. But I see that the Cambio Corsa did have a freewheel (similar idea). I was wrong to think it used four fixed gears. So indeed it could have used a chain tensioner!

Video showing it in action (skip to the middle): https://youtu.be/MQe0fLEQ2jY

I think there's a mixup between two things here. A _freehub_ contains both the cassette body and the _freewheel_. The former has been invented in the 1930s and popularized by Shimano in the late 1970s [1], the latter has been around since the 19th century [2].

[1] https://en.wikipedia.org/wiki/Freehub

[2] https://en.wikipedia.org/wiki/Freewheel

Just a link skipping to the gear change: https://youtu.be/MQe0fLEQ2jY?t=165

Yeah, I kind of let that one slide since it makes no difference from the standpoint of somebody trying to change gears while simultaneously not dying :-)

And let's be honest: it's entirely possible that the author has never changed a cassette or freewheel or seen a freewheel if they have. I'm almost 40, and I've only ever seen freewheels on pretty old 10 and 12 speeds (that's 2x5 and 2x6. Not a modern 1x drivetrain). And only because I volunteered at a place that fixes up bikes.

At the risk of this turning into an episode of Connections with James Burke playing out in text, why was there no freewheel at that point in time yet?

There was, the article mentions the cyclist having to pedal backwards in order to allow the chain slack to be taken up when gearing 'up' (smaller sprocket). That would be impossible without a freewheel, unless they mean to say 'exert some pressure backwards' rather than backwards pedaling, I don't see how a fixed linkage would be compatible with this system, it would behave terribly bad during a gear change because the system would always be under tension.

Freewheels were in use in the tour since 1936 at least, and possibly even earlier:

https://hizokucycles.bigcartel.com/single-speed-history

A cycling historian could clear up whether freewheels were allowed in racing during that time period. They were forbidden in the original Tour of France. Variable gearing already existed, in enclosed gear hubs, and remained a preferred choice for non competition cycling for at least into the earl 1970s.

Worldwide, enclosed-gear hubs are STILL vastly more popular than derailleur-style gearing implementations, btw.

They're part of what we might call the divide between "cycling" and "riding a bicycle." The cycling hobby is mostly about go-fast or go-far for its own sake, but in much of the world a bicycle is just transportation. In those places, you trade away efficiency or speed or weight in favor of the "it just works forever" bulletproof-ness that the enclosed hub provides.

Indeed, my daily driver is a Sturmey Archer AW from 1963. But I must say, contemporary derailleur systems with indexed shifting work pretty darn well.

They need maintenance that sealed-hub bikes can mostly do without, though.

My road bike has SRAM eTap on it. It's beautiful, sleek, and shifts perfectly almost every time. BUT that performance starts to suffer if I let the drive train get dirty, especially if I let gunk accumulate on the jockey wheels.

Your Sturmey can be ridden without such concerns. I mean, I'd never be able to keep up on my group rides on it, but for a general use bicycle it's probably superior.

I built a three speed road bike, and its downfall was riding with others due to the low ratio. When you're riding solo, you can ride at the right speed for a gear. But in a group, others are choosing your speed for you. Also, I like having a few more gears after mile 30 or 40.

Oops, not low ratio. I meant wide spacing.

Given a totally free choice, I'd love to try building up a racing frame with a modern hub gear. All those moving parts exposed to road grime, gone.

Well you can do it if you have the money. They're slightly less efficient than a regular cassette and derailleur, but if you're not racing it probably doesn't make much difference.

https://www.rohloff.de/en/products/speedhub

The gravel sub-genre has spawned a few frames built for the Pinion drive gearing solution (located in the bottom breaker area, like a central ebike motor, started by car gearbox engineers). Arguably that could be considered the next level of hub gears.

I have only personally had two serious failures of a bicycle transmission system, and both were on a rather new high end Rohloff hub. One was a mechanical failure (seal failure causing an oil leak). The other was a complete e-shifting failure caused by the most excellent engineers at Bosch’s failure to allow the shifting system to operate independently of the ebike drive.

(The latter is such a severe design error that I think no one should purchase the Bosch+Rohloff system until or unless it gets fixed. Use something else. An ebike is not a car, and it should be operable as a rider-powered bicycle even in the event of an electrical fault.)

Dropouts without chain tensioning slot were simply unheard of at the time and if you happen to be the inventor of the quick release it's not really surprising that you make it a key element of the shifting process.

I was wondering for a while about the purpose of the teeth on the axle/dropout, because clearly with bearings between cogs and axle there's no way to force the wheel forwards pulling on the chain (you force the wheel backwards for higher tension by applying a little braking force), but the teeth on the axle are what makes the movement in the dropout symmetric, so your wheel will remain aligned to the frame center line. Really clever!

This is the thing that seemed most crazy to me. They are actually adjusting the wheel in the dropout whilst zipping along in a scrunched over position. Seems like if you messed up couldn't the axle pop out of the dropout with horrendous results?!

Likely, if they didn't include some equivalent of "lawyer tabs" on the dropout. Compared to the general level of risk-taking of doing things like competitively descending Galibier or Tourmalet on wooden rims that added risk is probably a rounding error.

My father had one of those bikes when he was 20. He told me that he never managed to switch gear uphill like pros. He always had to stop, switch gear and start again. So it was not easy to master.

Reply on news.ycombinator.com