Hacker News new | past | comments | ask | show | jobs | submit login
Ceramicspeed chainless bike can nows shift gears and carry load (mashable.com)
86 points by jaytaylor 22 days ago | hide | past | web | favorite | 99 comments

Shafties come around about as often as airless tires — and stay just as long, too:

Saint Sheldon says “just use a chain case”, and as usual, he’s spot on:

Shaft drive was briefly popular around 1900, and occasional attempts are made to revive the design. Unfortunately, shaft drive turns out to have more problems than advantages.

A shaft drive requires heavier frame construction around the bevel gears to maintain their precise alignment under load. The drive system is heavier and less efficient than a good chain drive.

For reasons of clearance, the bevel gears of a shaft drive bicycle must be considerably smaller than the typical sprockets used with a chain drive. The smaller size of the gears causes an increase in the stresses on the whole support system for the shaft. This problem is exacerbated because the stresses from the shaft drive are not perpendicular the triangulated structure of a bicycle frame, and so are not well-resisted. .

Most of the advantages touted by proponents of shaft drive are only advantages compared with open-chain, derailer gear systems. Many proponents of shaft drive use specious (if not dishonest) arguments "comparing" shaft drive systems with derailer gear systems. Any such comparisons are meaningless, it's like comparing apples and locomotives.


Shaft drive proponents also often compare sealed, enclosed shaft drive systems with open, exposed chain drive systems. This is also a misleading comparison. All of the advantages claimed for shaft drive can be realized by the use of a chain case.


Shaft-driven motorcycles are a whole other story.

Are you talking about an enclosed derailleur system? I've never seen that.

Or are you suggesting we use hub gears? Because those are significantly lower efficiency. Only Rohloff is really competitive but it's still around a 5% efficiency loss compared to derailleurs.

> The drive system is heavier and less efficient than a good chain drive

They claim 99.2% efficiency in the video over 97% for the top-performing chain driven versions. Do you have any data on chain driven systems that are better than what they compared it to now?

Those aren't bevel gears in the video. Sheldon is talking about an older design that had cone-shaped gears which would suffer from the design flaws he mentions.

This shaft drive uses bearings for the contact between "bevel gears", making it a LOT more efficient.

So this new system saves about 2% of effort, which is large savings for competitive cyclists.

For 99% of cyclists, though, it doesn't matter. The overwhelming part of your work goes into air resistance, then rolling resistance at lower speeds, and probably the chain in last place.

In the last few seconds of video the mechanism can be heard -- oh boy, it is loud. The pawl in bikes can range from nearly silent to annoying, but this one seems to be even worse.

They claim 99.2% efficiency (in lab conditions) and compare that to 97% for a chain drive. In lab testing, bicycle chain drives have been found to be 98.6% efficient: https://web.archive.org/web/20080306054536/http://www.jhu.ed... so that would be an improvement of 0.6% rather then the 2% quoted..

And also most people's bikes aren't perfectly maintained.

Probably this system also loses efficiency if not perfectly maintained.

Looks like it has less exposed moving parts that a classic bike. Could be easier to maintain / clean / lube.

Not the naked variant — it's about the same as a chain derailleur. Maybe with the cover. It doesn't exactly look easier to clean that giant sharp dinner plate. Idk.

But perhaps for the average cyclist this "split pinion works under load shifting mechanism" is easier to use and more reliable.

To me it sound like it is possible to shift only one gear at a time and only when pedaling. So if you stop at traffic light with high gear, you can not switch to lower gear while waiting for the light to switch.

Most chain drives can't switch gears without pedaling either.

Internal gear hubs typically (always?) can change gears when not pedaling, derailleur gears can't.

And most chain drives are derailleur systems, not IGH, so you're both right :-).

And most IGH systems are chain drive as well, so all three of you are correct :-)

if chain includes belt.

My experience with Shimano gears is that if I change when stopped or while not pedaling the change completes when I start pedaling again. The only important thing is to go easy on pedals if starting from still and don't change both the front and the rear gears.

As a category of bikes that also can’t do this is quite popular this isn’t a deal breaking issue.

You definitely can, just hold the front brake and push the handlebars forward to lift off the rear wheel. It's even easier if you're using clipless pedals.

> For 99% of cyclists, though, it doesn't matter.

What matters is clean look. Chains are dirty and look dangerous. Good bikes (like Dutch omafiets) cover the chain, but if scaled manufacturing cost of this system is $30 extra, people would pick this one.

If I got one dollar for every time someone says the bike chain will be replaced...

Ya, the chain isnt leaving until something that can compete with it in a combination of weight, modularity, maintenance, cost, and performance.

Carbon belt drives are an alternative, but are not shiftable without internal gears like a Rohloff hub or Pinion gearbox/bottom bracket which are expensive and heavy, and regarding the Pinion, needs a proprietary bottom bracket built into the bicycle's frame. Belt drives are more much more expensive than chains, and also require a special frame, but they last much longer, and require virtually no maintenance. They are not nearly as efficient though which is why they aren't found on track bikes. Belt drives are great for riders that enjoy just riding along and want to invest in a nice maintenance-free alternative.

When Driven reaches production I would bet that it will go the way of the belt drive. It will have incredible improvements over a chain, but will be very expensive. I wouldn't be surprised if the total cost of the crankset, driveshaft, cog, and shifters are north of $3000 or more, not including the price of the wheel or frame. Pricing out almost everyone, except for the select few that find the improvements worth it. It won't be released until it is able to be raced professionally and I definitely think it will be seen in races where seconds matter.

It won't replace the chain as we know it, but this alternative will be on certain bikes owned by people that care about its benefits.

At scale production numbers, the cost should be much lower than 3000. And for bikes that aren’t as sensitive to weight, cheaper heavier materials can be used.

I believe it can replace the chain in all but the lowest bikes. It will start expensive, but so did many other technologies we now see on common priced bikes.

Belt drive does lend itself to CVT, so we might see a belt drive CVT bicycle someday, avoiding the need for an IGH.

Agreed. The bike chain is the thing that replaced shaft-driven bikes.


Motorcycles have had shaft drive, belt drive, and chain drive off the lot for decades. Many Harleys are belt-driven. Many BMWs are shaft-driven. Nearly all offroad and racing bikes are chain-driven.

It's not just friction efficiency, but weight and other factors, too.

The split pinion reminds me a bit of a DCT. Especially when they put the cover on.

Here's an off the shelf bike from 111 years ago with a shaft drive:


For some reason there's a tendency to assume that unusual things in the bike world are crazy new inventions.

I think the point was more that although other solutions exist is production, you won’t see them in high performance applications.

Both shaft and belt drives have benefits fir the consumer, but performance isn’t one of them. It’s probably similar bicycles.

Interesting, but this bike with a shaft drive doesn't seem to have gears.

Columbia did make a two-speed version. It’s been so long that I forget exactly how it works, but it’s not an internal hub.

Here, this page has a picture of the two-speed. Third down, on the left: https://www.prices4antiques.com/Bicycle-Pope-Mfg-Columbia-Sh...

Shaft driven bikes are not uncommon [1] [2]. This variant on the theme seems to be more about the novel gearing system than the lack of a chain.

[1] https://en.wikipedia.org/wiki/Shaft-driven_bicycle

[2] https://beixo.com/categorie/city-bikes/

I’ve been a cyclist all my life, including hanging out with antique bicycle clubs (I have an 1886 Columbia highwheeler) where one might, might, see an 1890s Columbia shaft drive. I might have seen a half dozen of the Columbias over the years. I have never seen a shaft-driven bicycle made in the last 100 years in anything but pictures.

So though shaft-driven bicycles most certainly exist, I would argue that they are the dictionary definition of “uncommon”.

It depends where you live. I imagine they're pretty uncommon in the USA. I've seen one or two in Australia. Most companies making shaft driven bikes seem to be in countries where bikes are used more commonly for commuting like Germany, The Netherlands and Scandanavia.

'not uncommon' seems a bit of a stretch, but I'm not surprised to learn (because they are uncommon enough that I've never seen or heard of them) that a Mashable post claiming something Is About To Revolutionize is nothing new, or at least what's new is more nuanced.

Looking at the design, I can't see how it would stand up to high torque or sudden accelerations. It would skip like crazy, and the flex on the frame would also disrupt the power.

I would guess this maintains its efficiency advantages only in very limited torque ranges and have overall power limitations that normal drivetrains don't.

For general transportation, electric bikes are the real revolution.

It would be great if they could get this into production. I like the idea of having the rear mech covered up form the elements.

Probably not for a race bicycle but this problem of 'covering from the elements' is kind of solved. a lot of tour bikes or city bikes 'stadsfiets' have internal shimano nexus gear hubs and internal hub brakes with an additional rain cover for the chain.

These bikes stay chained outside overnight for years with minimal maintenance.

Ugh, I had a 3-speed shimano IGH and while I loved the idea, in practice I could not shake the feeling that I was always pedaling uphill. That is to say, it felt like the IGH had considerable drag. Fun-killer.

There are probably nicer IGHs? 3-speeds are fairly low end. I don't know if the best IGH feel as good as chain drive bikes, though.

Yes indeed. Even my ~100$ bike has a cover and a basic Shimano gear. However, some parts of the chain are still exposed to the rain, but the oil makes the water slide along.

Internal gear hubs have been around forever.

If this drivetrain achieves its claims I will be very impressed. Shaft drive designs have been around for as long as chain drives; and they have never been as efficient. Much is lost in the torsion of the shaft. A 2% increase in efficiency is an absolutely massive gain. The cycling industry is littered with flashy looking innovations that don't live up to the hype. The basic design of a diamond frame and chain drive has not been improved upon for over 100 years.

But every now and then there are innovations that actually take hold. For example, the slant parallelogram dérailleur, hydraulic disk brakes, suspension (for mountain bikes), and recently single narrow-wide front chain rings. We shall see how well this drivetrain goes.

> The basic design of a diamond frame and chain drive has not been improved upon for over 100 years.

Except we have in the form of recumbent bicycles which have a huge aerodynamic advantage but they were banned in the 30s by making rules to enforce the frame dimensions.

Recumbent bikes hold the world land speed record on a bicycle if the definition of a bicycle is "2 wheeled muscle powered vehicle". For example the flying 200m speed record is 133km/h (flying start, flat ground, 1.66m/s tail wind max). 1h distance record is 90km while for normal bikes it is 55km etc.

Basically rules stifle a lot of innovation in the cycling world. Most pro bikes are too light to race legally so they have to add led weights. Strict rules about handlebars so you cannot get into a good aero position easily etc.

These rules apply to racing bikes, not to bikes that are used everyday for commuting or recreation. If recumbents were so much superior to the classic diamond frame, they would dominate the non-racing-bike market. I rode enough recumbents to prefer my uprights when commuting in a city, also for mountain biking and rides through hilly areas. If a was commuting in a flat area outside of a city I would maybe consider a recumbent.

A recumbent isn't "superior" to a traditional frame. It has select improvements, but for the general consumer the ride quality of a traditional frame is much better.

The general consumer has never experienced a recumbent ride. If they had, they would find the recumbent more comfortable on longer rides.

For bumpy roads, you can always stand on the pedals on an upright bicycle and use your legs as shock absorbers. You can't really do that on a recumbent.

But the general consumer most likely doesn't want to spend too much money on a bicycle and recumbents tend to be a bit more expensive compared to uprights.

> the flying 200m speed record is 133km/h

Todd Reichert achieved over 144km/h in a Sep 2016 record (almost flat; average down-slope under 2/3 %)


cycling competitions (as all sport competitions) are about the sport not the technical innovations behind it.

UCI regulations are there to make a clear separation between sport and tech. given the latest advancements in engineering it's possible to render the athletic abilities of a person irrelevant. power to weight ratio is an important stat in cycling and there is serious physical preparation involved in the process of dropping a few hundred grams in weight before a competition. now imagine that instead of that you get the latest innovation in carbon frame design and drop that weight from the bike. tech wins. and that's not the purpose of sport competitions. F1 has similar regulations for the exact same reason.

also, these regulations address health and safety. some aero positions can affect a rider's stability and, in turn, can affect the security of other riders.

As far as I know the diamond frame is mandated by UCI for professional races.

Lots of improvements have been made on this. Google 'recumbent bike'. I drive past a guy most mornings in one of these things [0]. He is driving that thing summer and winter, rain or shine.

[0] https://billeder.velostrada.dk/wp-content/uploads/2015/12/M%...

When someone wins an olympic event or whatever biking competition with this, we'll know if it's the real deal or not.

For the normal consumer, probably more about price than anything else, unless it enables better electric/motors, etc.

>For the normal consumer, probably more about price than anything else

This way just be the bubble that I live in, but I think the average cyclist that I know cares more about tech and marginal performance rather than price. New (performance) bikes commonly sell for thousands of dollars. It's not a world that seems especially price sensitive.

They won't because olympics and other biking sports are heavily regulated.

We've had bikes that can go over 100 km/h since decades and they aren't legal in biking sports because of these rules so bikers in sports only go about half that fast.

How does the rider trigger the shifting? The interviewee refers to the system as "wireless", and I assumed that meant "no shifter cable" (as opposed to "using electromagnetic energy for communication" or something). I can't quite see in the video what they're doing to cause the shift up or down to happen.

In bikes these days, they do indeed mean "using EM energy for communication," AND also "no shifter cable." There's a comm link similar to Bluetooth (or maybe straight-up consisting of Bluetooth, I don't actually know) between the shifter controls and an actuated derailleur (or whatever the shifting mechanism is called in this case).

Probably something lower power than BT. These things try to have days of battery life with tiny batteries (e.g., Di2, eTap).

Modern Bluetooth can have years of battery life on a cr2032 button cell if all it does is signal some inputs.

What kind of IC do you have in mind for that kind of lifetime? While signaling, say, twice a minute on average, 2 hours a day?

I have no idea how it’s built but I’ve seen it work in practice.

This kind of device works for years on a small battery, signaling every rotation of your wheel to your phone as you cycle:


The spec page says 500 hours of battery, which is about 20 days. At 2 hrs/day that's 240 days -- still quite a bit, but yeah.

I thought I saw a serial cable port on the shaft - I'm guessing it's wireless and battery powered?

Oh good, I can’t wait until a Ukrainian teenager decides to shift my bike down eight gears in traffic for the lulz

eTap baby. This tech has been around since 2015:


Thanks for the link, I have definitely learnt something. I didn’t suspect such system to exist outside a Sci-fi book!

Sounds like it might be pretty loud

I'm really looking forward to more belt driven bicycles with internal gearing like the Priority 600 and Continuum Onyx. I'm tired of delicate bicycles that I have to fiddle with CONSTANTLY to keep them running.

Buy a better bike. You should effectively never need to fiddle with anything if you spend $300+ and aren't riding lots of hours or through intense terrain.

Yeah I don't see much smooth roads and ride about 500+ miles a month. Some of it is on roads, but also a lot of gravel/dirt. I have to constantly clean chains and group sets otherwise shifting gets super crunchy.

I used to ride a cheap bike (Liv Alight 3), but after replacing 3 wheels it was cheaper to buy a nicer bike. Tourney has become a swear word in our house. Lately I've been riding my Catrike Expedition which is a blast, but chain and gear maintenance is still a problem.

Sounds like you might be well served by a fat tire gravel bike, 1x up front with a limited number of rear gears? The less gears, the less fiddly, and they are built for rougher road.

2% less effort to ride is fine and all, but how easy is it to maintain? I'd pay extra for a bike that never needs maintenance – does not need to be oiled regularly, does not need replacement gears or chains etc.

When I hear "novel" and "patentable" mentioned in a video, I'm quite sure it's made solely to impress investors and immediately assume most of it is BS.

How is it that you're sure?

I am not a fan of Patents, especially software patents and ones that are deliberately filed for malicious, anti-competitive measures.

When a company truly innovates, they deserve the prize of a patent. It appears that they've done so, investors lurking around or not.

I think the point of GP is more that their marketing content seems to be geared to wow investors, not consumers. Mentioning that they have a patent is of no positive value to consumers, it in fact might suggest that the design will be a niche commanding a high price.

They specifically called out one particular aspect of their design as being novel and patentable, about halfway through the video. Shaft drives for bicycles in general are not new. Calling out that one particular feature this way tells the consumer watching what sets theirs apart.

Let them, it is better to get funding from investors if you actually have a novel idea. What's wrong with marketing to investors? Unless you have some shady solar roadways idea or a way to produce water out of thin air, that is. May be I am getting older and less cynical.

This is an interesting article about drivetrain patents. Shimano, Campagnolo, SRAM, etc.


Who wants to have an open source cycle gear system. I'm in.

But here is what I think it looks. Whole concept of patent world is flawed and here is how it is.

This piece of gear system is a mix of central shaft(that we see in trucks + gear + that cut-in-half gear tooth)

If we split it into three individual item, then it is a three different patented item. Put together you get a new patent? Come on!

Reminds me of those old hand crank egg beaters.

Motorcycles have had belt drives for decades. I wonder if they'd help bicycles also. (Lighter, cleaner, etc.)

...considerably less efficient. Doesn’t matter in a 100+ horsepower motorcycle. Matters a lot on a < 1 horsepower bicycle. Pain in the ass to change, too.

Previously, sharing bicycles company (Mobike) in China used similar chainless bikes in their businesses for ease of maintenance. But they were later replaced by traditional ones because chainless bicycles were usually much heavier and consumed more energy from riders. Very curious about how people in the article solved this problem.

The article on CT gets into much more detail about this system: https://cyclingtips.com/2019/09/ceramicspeed-driven-can-shif...

tl;dr: interesting and shows promise, still more work to do.

edit: don't be scared by the not-a-paywall greyed out text and box.

Interesting idea for shifting. One thing I'm concerned about (and this is probably only me in the entire wide world) is the tactile feel when shifting (and, indeed, pedaling in general.)

On single-speed bikes and common derailleur type multi-speed bikes (yes, even with freewheeling, but I assume even more without it) there's a very direct linkage between pedaling and movement. You start moving pretty much the instant you put pressure on the pedals.

I've noticed on some hub-gear bicycles there's a significant amount of slack where you have to pedal without resistance for at least 1/8 of a revolution until the drive mechanism kicks in. This is very frustrating to me (again, I'm sure it's only me...) and completely disrupts the suspense of disbelief I otherwise can conjure, which seems to be what allows me to get that freedom feeling almost like I'm flying. The lack of feedback for that fraction of a second yanks me straight out of the flow of navigating traffic with the bike as an extension of my body and reminds me sternly that I am sitting on a very-separate-from-my-body mechanical horse, some aspects of which are very much not in my control.

That sense of freedom is definitely one of the major reasons I use the bicycle as my primary means of transportation.

Sorry, very long rant about something very unspecific but vaguely related.

A 2% bump in efficiency hardly seems like a revolution. This is still cool though, I look forward to bikes that need lower maintenance because of the lack of chains

On the one hand, it's a 2% gain in efficiency. On the other, it's a 60% reduction in losses. From that viewpoint, it's pretty significant.

Drivechain losses only, and only if you take their numbers at face value. And if you ignore losses due to shifting. There are plenty of other losses (aero drag ...).

If UCI allows it on race bicycles and I if it can survive real roads everybody is going to use it in races no matter the noise. 2% is huge.

For amateurs doing races, it's the same. For the huge majority of amateurs it depends on cost, durability and ease of maintenance. For everybody else, it's only the cost.

Can you imagine taking a dinner plate of shark teeth rotating at ~20 Hz in the calf? I kind of doubt UCI would allow this on race bikes, at least uncovered.

I presume 2% efficiency means is a difference of many seconds in competitive cyling. It is huge. When I used to work at GE Gas Turbines, a 2% efficiency increase from generation to generation would destroy the competition (Siemens). It could mean millions of dollars of profits for energy companies that run these turbines 24/7. Not saying its the same thing, but 2% efficiency increase in most highly saturated markets such as LEDs to cyling to wind turbines - is HUGE.

I’m interested in the idea of going not just chainless, but not having any rigid or semi-rigid drive connection at all.

I ride a recumbent tricycle, and an interesting complaint from my dad when we were discussing why recumbent trikes and velomobiles aren’t more widespread is that the design is unweildy, that the end result doesn’t look polished; so we discussed ways that you could make it neater. The chain on recumbent trikes goes from the very front to the back of the vehicle, so it’s a long, subjectively unsightly thing. Something like the this Driven shaft drive would be quite unsuitable for a trike due to the length of the shafts that would be required, and the resulting torsion.

So then, we wondered, what would it be like if you replaced the entire drivetrain with an electric drivetrain? Pedals at the front would drive a generator, which would pass power through slim wires hidden inside the frame or similar, to a motor at the back. Much neater, visually if in no other way. This would fit in very nicely with making it an e-bike, too: precise legislation wording could be problematic, but your 200W or 250W or whatever would be added to whatever power you generated by pedalling. And by being a fully-electric drivetrain, efficiency would be improved on the electric assist. (Probably similar to the benefits that I understand going full-electric on hybrid car drivetrains can offer.)

I’m not deeply familiar with the efficiencies involved, but my impression from light research is that regular chain drives start out at up to 97% but often operate at figures closer to 80–85% due to wear and lack of maintenance; and a full electric drivetrain could be over 90% efficient, though it’d be more likely to be around 80%. (Remember, we’re not getting a battery involved at this point, so no losses on that.) Perhaps such a drive would be on par with a typical chain drive for efficiency, though less than a well-maintained one; but it could provide improvements in efficiency in other areas (electric assist and foot-to-bike power transference).

I believe that less maintenance would be required, and it’d the complaint of the dirtiness of bike chains too. Heavier parts and more expensive, unless you were going e-bike already in which case it may be lighter and cheaper, which would become a much more obvious thing to do for such a bike.

It’s interesting to think about. But I doubt I’ll ever do any practical experimentation on the idea, because my list of projects I’d love to work on is already more than thirty years long (and getting longer each year), and this sits a fair way back on it.

Hydraulics would also be interesting to experiment with. https://en.wikipedia.org/wiki/Chainless_bicycle is a good page to start on for learning about all these things.

As a slightly related precedent, some early trucks used an engine->generator->wires->electric motor system as an electric transmission as this worked better than the early gearboxes.

On a bicycle, the wiring would certainly be easy and considerably lighter than shafts, chains or hydraulics. You would take a hit on efficiency as pedaling force would have to be converted to electricity and then back to motion, but generators and motors are pretty good these days. Weight could be an issue, but again the developments in hub motors have seen the weight come down a lot to the point that your idea seems viable.

The efficiency of electric assist doesn't really matter. They get decent range with a small battery, so you can just use a modestly larger battery to get more range.

I'm not saying more range from the same battery would be worthless, just that the ranges they already offer are plenty marketable.

I also wonder about the efficiency of a small (tiny!), slow generator.

You’re generally correct, especially with how people mostly use bicycles and e-bikes. But it depends a bit on how you’re positioning it; two interesting cases that I’m thinking of: using a “trickle power assist” to cheat your way to effectively 100% efficiency, so it needs to be efficient with small augmentations from a battery (which I think is the sort of case that would commonly be particularly inefficient, but when combined with your pedalling in the same drivetrain I imagine it might not be); and velomobiles for longer-distance travel (very much more being a car replacement), perhaps even touring, where larger range figures start to matter more. If you can get it to go 20% further at a given speed, that becomes a bigger deal.

A substantial art of this thought experiment is about designing a delightful product. As my dad said to me: if Apple were designing a bicycle or tricycle or velomobile or similar, what would it be like? We both agreed that there wouldn’t be an exposed chain, for starters, and I suspect that they’d at least try to do away with the chain altogether, and they’d care about efficiencies if they went down this path.

I thought it went against HN policies to post about startups that actually follow through on their pie-in-the-sky ideas?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact