CeramicSpeed’s Driven Concept Might Become the Most Efficient Bicycle Drivetrain | Hacker News Reader

CeramicSpeed’s Driven Concept Might Become the Most Efficient Bicycle Drivetrain

bicycling.com

95 pointsuntangle113 comments

As mentioned (and slightly buried) by the article, the really fascinating part is that we already have consumer bicycle gears that deliver 97% efficiency: a Dura Ace groupset is expensive compared to the stuff you'll get at Wall-Mart, but they're sold at pretty much every cycling store. I wouldn't be surprised if the Ultegra and 105 (Shimano's next two groupsets by price) were nearly as efficient as well.

2% is nothing to sniff at in cycling (people pay thousands of dollars for that kind of advantage), but there are other factors as well: resistance to stress (sprints), ease of maintenance and service, and weight all factor into the utter dominance of the current groupset design. It'll be interesting to see if CeramicSpeed can advance their design on those fronts.

Can you run torque through two 90° angle changes for less than 3% losses?

Depending on how you do it: yes. Look closely, the gear teeth on the driving wheel are miniature roller bearings.

Are you sure those aren’t bushings? The round bits in a bike chain are bushings, which is part of the efficiency of the existing drivetrain.

These folks have essentially turned the chain inside out and attached it to the ends of a shaft.

https://hips.hearstapps.com/hmg-prod.s3.amazonaws.com/images...

Look closely in that picture you can clearly see the inside of the small bearings.

Well I'll be damned. Them's bearings alright.

That's also a weak point: those tiny bearings are going to take all the load from the cyclist one or two at the time, I predict they won't last long if it works at all under substantial load.

And to reduce friction they're open cage which will not play well with sand and other grime.

Possibly. They must be getting their idea out 99% from somewhere, and existing transmissions in other application domains are a likely candidate. But I really don't believe that it could work at bike-friendly torque/weight ratios, in a system open to the gritty environment, in a multi-gear setup where angles are not aligned (the inner part of the teeth moves slower than the outer part, permanent slippage, yay).

As far as I understand it this is a mockup celebrating the intention of getting to 99, not a prototype that has achieved it. More artist's impression than product.

What's still impressive though is that the concept does away with pretty much the only actual product the company has, the pulley wheel. That is bold in a strictly economic way.

> we already have consumer bicycle gears that deliver 97% efficiency: a Dura Ace groupset is expensive compared to the stuff you'll get at Wall-Mart, but they're sold at pretty much every cycling store. I wouldn't be surprised if the Ultegra and 105 (Shimano's next two groupsets by price) were nearly as efficient as well.

There isn't really much difference in efficiency with price in group sets until you get into the really bad cheap stuff. Most of the differences are cosmetic or ergonomic and status related. Chains and bearing and cogs are the only items that affect efficiency and they are all pretty good. Good chain line and proper maintenance and lubrication make more of a difference.

Internal hubs are slightly less efficient overall and this depends a lot on the internal design and which gears are in use. But even these are something like 94% - 97% efficient.

Shaft drives have been the "next thing" since the end of the 19th century but have never managed to combine reliability and efficiency well. I doubt an exposed mesh 90 degree gear set with what look like teeny tiny exposed ball bearings are going to be as efficient after a good splash of water laden with brake dust and road grit. It might work if they enclose the whole system.

weight is a big factor between cheap and expensive groupsets

Overheard at a popular stop for bikers near me: "Yeah, I could have gotten the model above it which is 500g lighter, but it would have cost me another $3500."

It should be noted that the guy who said it could have easily lost 500g by a minuscule adjustment to his daily snack intake...

No, it should not have been noted.

Maybe it's a local thing, but a considerable number of the guys stopping there (it's pretty much the midpoint on a quite nice and highly popular round trip) are middle-aged, and clearly quite well off. So they eat well and ride expensive bikes...

I'll just leave it at this: this particular guy's eating habits, whether you're privy to the details of them or not (and I suspect you aren't), were totally not germane to the discussion. The first sentence was legit, and the second was entirely unnecessary.

To be fair, this is in fact a very common observation among cyclists, with no snark intended. Obviously, mass is mass, it matters not one whit if it's meat or space-age metal.

It's just funny how many wannabee Eddy Merckxs seem immune to that truth, and we who grok it like to snicker when we hear discussions like the one cited.

Not quite. rotating mass (e.g. tyres) is more important to bike handling and efficiency.

I said snack intake because most people do eat some snack, and it's the easiest area to cut calories. Of course, he could alternatively have cut down his dinner portion by a slight amount for the same effect.

500 grams is ~10 calories per day for a year, and 10 calories is 1/25th of a Snickers bar or about one eight of a potato.

Either way, he'd get the same effect has the extra $3500 would have, which was what I found amusing and why I mentioned it.

As the cycling component equivalent of the CAP theorem goes: light, cheap, strong - pick any 2.

How light, how cheap, and how strong, though? Seems like an incomplete theorem.

I had a similar experience: I could buy a steel water bottle cage for $12, or a nearly identical model made of titanium for a savings of a few dozen grams at the cost of an extra $60 or so.

This seems like an exceptionally silly expense for a device which is designed to carry two pounds of water around!

The "classic" Mobikes have enclosed shaft drives. But they also have airless tires, a tiny frame and crank size and a frontwheel powered LED system. It's hard to say where their perceived huge inefficiency comes from.

> Internal hubs are slightly less efficient overall and this depends a lot on the internal design and which gears are in use. But even these are something like 94% - 97% efficient.

This isn't true. Rolhoff hubs can approach that for some gears, but bikes with them start at $2K. Nothing else except derailleurs are even close:

http://bikeshed.johnhoogstrate.nl/bicycle/drivetrain/shimano...

https://www.cyclingabout.com/speed-difference-testing-gearbo...

Also, current shafts and bands reduce speed another few percent.

A well maintained Sturmey Archer three-speed will match the efficiency of the Rohloff, albeit with a far narrower range of ratios. They're an excellent choice for utility bikes in relatively flat areas.

Wealthy middle-age people and pro now spend fortune to save half a watt so that 2% gain is indeed not marginal at all. Although I tend to be very cautious with manufacturers claims, for some reason especially CeramicSpeed. That being said, I wish them to succeed, as it will bring some fresh air to cycling.

>2% is nothing to sniff at in cycling (people pay thousands of dollars for that kind of advantage)

2% is nothing to sniff at only if you're hampered by the arbitrary UCI rules. By far the biggest factor in cycling performance is air resistance, and the obvious solution (fairings) is banned. The UCI has been disastrous for bicycle technology. Most high-end buyers like to pretend they could compete professionally one day, so they abide by the same rules, which means there's no incentive to develop truly fast bikes.

The UCI rules should allow any safe design, and avoid giving an advantage to richer teams by setting a price limit (bicycles are already required to be commercially available).

I agree that the UCI has been bad for progress. That being said, I raced in college (at the collegiate level, not professionally) and saw plenty of people using UCI-noncompliant bikes.

Historically, the UCI did use "safe design" as the metric -- they used it to exclude Graeme Obree, whose bicycle-cum-dishwashing-machine was probably safe enough for him to be riding on a closed track. That's not to say that the UCI should block innovative designs, just to note that they'll use any regulation supplied to restrict cyclists/companies/countries that end up on their naughty list.

Do you think there's a market for super-efficient commuter bikes? Or is air resistance no longer the limiting factor at commuter-bike speeds and weights?

I think there's a limited market for them, because the value of aerodynamics depends heavily on how you ride.

Wind resistance increases with the square of the speed. Pros can ride at 30mph average on fast stages, where managing wind resistance is huge. A typical road bike commuter might average 15mph. All else being equal that's 1/4th the wind resistance (1/8th the power needed to overcome it), which is still big, but not the overwhelmingly important concern it is for the pros. A commuter on a utility bike who doesn't want to sweat might average only 10mph, so 1/9th the wind resistance (1/27th the power needed to overcome it). At that speed it doesn't feel so important.

For the pro, speed is the top priority, but the commuter also has to consider comfort, reliability, and visibility of other traffic (recumbents put your head lower down so it's harder for you to see over cars, and it's harder for them to see you). On heavily congested roads, e.g. central London, traffic might prevent you from going fast enough for wind resistance to be a major concern. If you commute over steep hills then reduced weight might benefit you more. But if you have a long and flat commute with light traffic, I think a super-efficient commuter bike would be valuable, and it's a pity that UCI rules discourage development of super-efficient bike technology.

EDIT: added power values as per HankB99's comment.

This is the kind of reasoning I was looking for. Thank you!

>Wind resistance increases with the square of the speed. To be pedantic, wind resistance varies with the cube of the speed. That's my recollection and seems to be used in the formula used to calculate power required at https://www.exploratorium.edu/cycling/aerodynamics1.html.

Drag force does vary with the square of the speed, but power needed to overcome that drag force varies with the cube of the speed. From Wikipedia[0]:

"Note that the power needed to push an object through a fluid increases as the cube of the velocity. A car cruising on a highway at 50 mph (80 km/h) may require only 10 horsepower (7.5 kW) to overcome aerodynamic drag, but that same car at 100 mph (160 km/h) requires 80 hp (60 kW).[16] With a doubling of speed the drag (force) quadruples per the formula. Exerting 4 times the force over a fixed distance produces 4 times as much work. At twice the speed the work (resulting in displacement over a fixed distance) is done twice as fast. Since power is the rate of doing work, 4 times the work done in half the time requires 8 times the power."

Power output is directly felt by the rider, so it's the more relevant comparison here, and your comment is more practically useful.

[0] https://en.wikipedia.org/wiki/Drag_(physics)#Power

Oh... right. Work done is force * distance so distance varies linearly with speed.

Fairings aren't as useful as you might think in most circumstances. On an upright bike, they need to be enormous to offer an appreciable benefit, which makes them heavy and difficult to control in crosswinds.

The optimum solution is a recumbent bike - by going feet-first, you can reduce your frontal area by more than 50%. With a much smaller frontal area and a much lower center of gravity, a tail fairing becomes a practical proposition.

Fully enclosed recumbents can achieve phenomenal results in the right conditions (Sam Whittingham's 91km hour record, Andy Wilkinson's 41 hour LEJoG), but they immediately become a handicap with any sort of gradient because of the ~20kg weight penalty and they're unbearably hot.

> The optimum solution is a recumbent bike

Recumbents are cool and much faster in some situations but pro cycling exists mainly to sell stuff and upright bikes look cooler.

Recumbents have been banned by the UCI since 1934, so that ship has sailed. If we had seen recumbents on the Tour de France for the last eighty years, they'd probably look cool and upright bicycles would look weirdly old-fashioned.

There is also the point that you don't want to ride a recumbent bike on a road with traffic.

Ok flat ground or a descent a recumbent may be faster (my experience is that most recumbents can’t keep 22-25mph on flat ground) but it won’t be faster climbing hills and they have a longer wheelbase and are quite a bit less maneuverable. On hill climbs the weight is the most obvious factor but I’d be surprised if you can generate the same amount of power in the supine position as you could in the standing position.

Since most bicycle racing is dominated by climbing hills, I don’t think anyone would ride a recumbent even if it was legal.

Why? Is there some physiological factor? Blood flowing too much to the head? Allegedly the hydrostatic pressure is lower, breathing is easier due to lack of bending on a recumbent bike, so it should help endurance, as long as the bikes are the same weight.

A modern short wheel base recumbent has a similar wheelbase to a conventional 700c road bike. The hill climbing issue is still the subject of some debate; your peak torque is undoubtedly lower (because you can't put your full weight on the pedals) but there's no obvious biomechanical reason for lower power.

Any advantage an upright bike might have in the mountains is completely neutralised by the tremendous aerodynamic advantage of a recumbent. A good lowracer recumbent with a tailbox has about half the drag of a normal racing bike. Worse still, an upright rider gets almost no benefit from drafting a recumbent rider.

If the UCI legalised recumbents, a team using upright bikes wouldn't even finish the first stage of a grand tour within the time cut-off. Any team using recumbents could disappear off the front of the peloton and would be impossible to catch. In a time trial, there's simply no contest - the recumbent riders could demolish the upright riders without breaking a sweat. To neutralise that advantage, races would need to be organised exclusively in the high mountains with no appreciable amount of flat roads.

Try making a bunny hop over a pothole on a recumbent. Upright bikes are far more versatile, just imagine recumbent cyclocross.

    > The optimum solution is a recumbent bike...

On a flat individual time trial, sure.

But pro races are really all about bursts of acceleration as well as topography. Recumbents just wouldn't cut it.

Old guys going fast in a straight line by themselves in an egg-shaped container just isn't interesting to watch.

    > ...which means there's no incentive to develop truly fast bikes.

UCI compliant bikes are "fast enough" for the sport and have many practical considerations in addition to normalizing design features.

It really is mostly about the training, the athletes and the teams.

In cycling, performance is result of the combination of many factors of which equipment is but one.

... ease of maintenance and service ...

Exactly, and gears and chain are horrible in this regard. The absolut vast majority barely maintains their chain which leads to a much lower efficiency and really poor longevity (no it is not hard, but apparently too hard/boring to get done).

Unless the owner is a bike enthusiast you can almost guarantee that the maintenance is abysmal.

I realized this for myself and bought a belt-driven bicycle so I wouldn't feel bad abusing my current bike. For your day-to-day bike I'm very happy with it and the maintenance benefits alone make it so worth it.

I use a single speed bike at the moment. Not because I hate gears but because I’ve never had a geared bike that didn’t constantly click on the outside gears. It drove me crazy trying to tune them perfectly and never succeeding. Forget increased efficiency. If a new design emerges that accommodates gears and is simple to tune properly, then sign me up. No idea if the design being discussed could offer this.

I only ride fixed so not an expert on groupsets but people using upper end shimanos does not seem to have that much tunning issue.

maybe you yours is damaged or sth?

Or mismatched. It's not rare to see people upgrade their bikes with newer gear but to end up with grips that are set for 8 or 9 speed and cassettes and derailleurs with more speeds. That tends to get very awkward.

Have you tried out a bicycle with an internal hub gear? I don't know how loud they are, but you shouldn't have any chain binding issues (and you could probably run a belt if you wanted).

I use a Gates belt drive with a Shimano hub gear and it’s nearly silent compared to a chain.

Oh man, I had a Ghost bike with a Continental drive belt that was a constant pain in my ass until the belt snapped at 1500 miles and practically chucked me in front of traffic.

REI really stepped up and Ghost warrantied the entire drive system as the Continental system was NLA; they replaced the whole thing with a Gates Drive system which is amazing.

I have not. Never even heard of that but I'll check it out now. Thanks :)

i've ridden single speed exclusively for ever a decade. first as a commuter, then fixed in a velodrome, and now freewheel in the mountains. i recently built my first geared bike since my first "real" bike i got as a teenager. it uses shimano xt di2 2x11 on onyx hubs. it's absolutely silent except for the shifts, which sound robotic. it's pretty fantastic. it was the only way i could get along with derailleurs, as i experienced the same problems you mentioned above. my commuter now has a shimano nexus 3 igh, and that's pretty great too.

Most indexed systems work just fine. Di2 is nice but a luxury imo, and the parts are just crazy expensive.

Electric shifting is at this point in time more of a status symbol than an actual solution.

Not if your bike gets serious miles on it. The rate at which I got through gear cables and outers before DI2 was quite extreme. To keep good shifting I'd have to change the both at least every 2 months, and ideally do a strip down and deep clean on the derailleur.

Not to mention the constant adjustments of cable tension as the cables stretch.

With DI2 I've not adjusted or stripped down for a good 2 years and I still have perfect shifting.

That's an interesting observation. What conditions do you ride in?

I cycle a lot (more than anybody that I know personally except for my brother) and tend do do years with cables and shifters. Cable stretch requires the occasional twist of the tensioners but that's no big deal.

I can see how for you Di2 is not a luxury though. Really curious about your mileage and other contributing factors.

Out of interest, what derailleurs and cassette's have you used?

I'm running a 10 speed Shimano XT cassette and an XTR derailleur on my mountain bike and it's amazing. Doesn't really make any sound and the shifting is downright amazing.

Its been awhile and I can't really remember. I spent about 700 on the bike so I thought it would have decent components. Maybe you have to spend more?

USD I presume, mountain bike or road bike? Because if it’s a mountain bike that probably would have been just above entry level.

Yeah, they were mountain bikes. Seems crazy to me that you have to spend $1000+ just to get something usable. Ah well... :P

If you’re not bother by efficiency then why not consider a hub gear system?

Rohlofff. https://en.wikipedia.org/wiki/Rohloff

Good stuff but expensive.

The cycle-around-the-world types swear by them, they are very reliable.

DAG-2.2 to the rescue.

https://www.parktool.com/product/derailleur-hanger-alignment...

> It drove me crazy trying to tune them perfectly and never succeeding.

Don't electric shifting solutions like Shimano's Di2 or SRAM's eTap system address this? At least I thought those systems also cover the calibration and fine tuning of the rear derailleur.

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. So mush 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 acually take hold. For example, the slant parallelogram derailleur, hydraulic disk brakes, suspension (for mountain bikes), and recently narrow-wide front chain rings. We shall see how well this drivetrain goes.

Shaft torsion does not lose you much energy, it mostly acts as a torsion spring: whatever you put in you will get back. The shaft will only warm up a tiny little bit from the amount of energy that it will lose during a single cycle.

Suspension does lose you a lot of energy, which is why you'll never see it on road bikes.

Springs are terrible when driven by biological "pistons". Applying a certain amount of force is much harder in some phases of the pedal stroke cycle than in others and springs would make it impossible to go easy when you want/need to.

It is a very hard spring, not something squishy like front fork suspension.

The biggest downside would be eventual wear of the driveshaft, the losses will be very low.

Note that almost every part of your bike acts as a spring in that sense, the frame flexes a bit when you pedal, as do the cranks and the shaft. Even the spokes in the rear wheel act as springs transmitting the force from the hub to the rim (which is one reason why they are oriented the way they are, that way they pull the rim along rather than that the spoke gets bent, the spoke is stronger in that direction).

Everything acts as a spring, and performance bicycle engineering goes to great lengths to make them as as hard as possible. Frame flex under pedaling load is measured, optimized against and a driver of buying decisions. Replacing the chain with something more springy? Good luck in that market.

> Everything acts as a spring, and performance bicycle engineering goes to great lengths to make them as as hard as possible.

Indeed. So if you use a shaft to drive the rear wheel that would definitely be part of the equation, I note they are using a hollow carbon fibre tube, which in that particular dimension is likely not ideal for the application unless it is given some more cross section. Even so, it is an interesting development.

> Replacing the chain with something more springy?

Chains stretch quite a bit, you'd be surprised.

"Chain stretch" is usually slang for chain wear. The chain gets longer as the rolling elements wear down so people call it "stretch" even though the metal isn't stretching. The actual stretch under load is very small.

I'm aware of the difference, but thank you anyway.

You can see the effect for yourself if you lock your rear hub and proceed to push down on the pedal (you can see it because the pedal is a nice long indicator effectively multiplying the distance the chain stretches).

A bit nicer setup is a micrometer at the end of a fixed section of chain with a weight attached.

Chain does stretch. About 1.5 mm under full load.

That's why you want chains with solid pins and solid plates.

What you are talking about is chain elongation as a result of wear, essentially the accumulation of slop in the bushings the pins go through.

Yeah, shaft drive bicycles ("chainless") were marketed as the "next big thing" around 1898–99 by big companies, like Pope Manufacturing Company's Columbia brand, as they were coming down from the Bike Boom of the 1890s.

The important thing with bicycles, compared with motorized transportation, is that the "engine" is still the same now as it was then, so everything is very limited by weight and efficiency. We wouldn't be able to practically use most motorcycle and automobile improvements that have come since the Model T if we were stuck with the Model T's engine power.

And FWIW, I ride a motorcycle with a shaft drive. The fastest motorcycles all use chains. Many now use tooth belts, but I think they may be limited in the power they can transmit.

Check these out: https://pinion.eu/en a german startup which created a gearbox for bikes which works like the one in a car. Early Investment came from an engineer at Porsche.

Disclaimer: not affiliated

Unfortunately, Pinion systems haven't really gained any significant market share due to a) really high cost b) inability to downshift under load, c) a 1.5lb weight penalty and d) the frame must be built specifically for the Pinion system[1]. Still, always great to see new approaches like this as bike maintenance is a huge barrier to adoption.

[1] https://www.singletracks.com/blog/mtb-gear/the-pinion-gearbo...

Very nice concept, I really like the fact that the casette is gone because it occupies a lot of space leading to substantial weakening of the rear wheel (the width of the hub where the spokes attach is a very larger factor in rear wheel strength).

That gear does look like something that would do well in a meatgrinder, and given the fall-out over just having disc brakes on racing bikes I don't think that would pass inspection for road bike racing.

10 points for out of the box thinking though, a cardan driven racing bike is very clever.

I cardan shaft just another name for a drive shaft?

https://en.wikipedia.org/wiki/Drive_shaft

Sorry, probably a dutchism...

We use the same term for the rear end of a car and for the drive train on BMW motorcycles.

Yeah, that's going to skip under load and won't stay aligned.

Might be usable as an electric assist drivetrain and it was sealed up.

Yes. Based on that photo, I can't imaging that the set of ring gears could be rigid enough. Unless it was inside a strong casing. As in standard differentials.

This is cool and really fascinating, technically, but ultimately a solution to a question nobody asked and I don't see many benefits outside of a claimed efficiency benefit while there are quite a bit of drawbacks. Also I'm failing to see how this is different from shaft driven bikes [0] we already have had for 'some time' now

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

The system would have to be pretty intelligent to know how fast the system is moving. It’d have to be pretty smart to know how fast it’s moving, which tooth track to select to make the shift happen. Would you have to back off to shift under high load?

This reminds me of some newer automatic transmissions that use dog clutches, like a manual one, relying on sensors and electronics to do the synchronisation --- there is a very noticeable (and unpleasant) jerk in the shifts, since it has to match the speeds precisely, and automatically reduces throttle to do so.

Of course, with a human providing the power directly, that's not really possible. I suppose you could add a "shift light", but IMHO that's just overcomplicating things...

Other options include CVT which offer continuous non-integer gear ratio steps, but a lower efficiency of about 79%.

http://www.enviolo.com/nuvincicycling/nfinity

They mention something about an electronically assisted shifter. Doesn't that defeat the whole point of a (non electric) bicycle?

Electric shifting is present in high performance cycling for almost a decade now (more if you include failed attempts like the Mavic Mektronic from the 1990ies).

It's a roughly 50% split between users who want electronic shifting and users who don't (personally, I don't, I'm pushing more than my share of buttons when I'm not on the bike, I find the refined engineering of a mechanical groupset far more impressive than just throwing microcontrollers at the problem)

Not really. All three of the major manufacturers offer electronically-controlled derailleur gear systems. It offers minor aerodynamic benefits and more reliable shifting performance with no real weight penalty. The battery adds about 60 grams and the derailleurs are slightly heavier, but the shift levers and cables are significantly lighter. That battery lasts for about 1000 miles of typical riding, which is more than enough for the target market of racing cyclists.

The inconvenience and cost of electronic shifting isn't a good fit for leisure and utility cycling, but it makes sense in racing where every marginal gain in performance is valuable.

Ah well, when I think cycling I don't think racing, I think out-of-town rides where it helps if you're able to fix your bike using just some wrenches...

My grandfather told me about a bicycle he designed that uses a CVT. Curious if that’s a better direction for the future of bicycles.

The problem with CVTs is that they're relatively inefficient.

You can buy these today, and many e-bikes use them. About 15% power loss means that for an un-assisted bike they are not an option.

(Dutch link: https://www.fietsenwinkel.nl/expert-e-bikes/nuvinci with nice cut-out picture of how it works)

Unproven, can't yet change gears, based on an old design used over 100 years ago that was given up on.

> based on an old design used over 100 years

Sounds interesting, can you point to some documentation?

> A stock Dura Ace drivetrain returned about 97-percent efficiency.

Does anybody know the numbers of efficiency in fixie and single speed bicycles? I reckon they should be better given the absence of a derailler? I tried to search for it, but couldn't find numbers.

Another article on this technology stated that the efficiency gain comes from there being only 4 points under load versus 8 points in a chain with derailleur system.

A fixie also has only 4 points, since the derailleur is gone, so it's presumably more efficient.

But the big problem is that there is also an efficient human cadence (I think around 100 RPM), and a fixie has a massive problem here. So overall, a fixie is massively more inefficient. Which I guess is no surprise, cycling competitions are run with geared bikes, not with fixies.

This doesn't sound correct to me - the derailleur is on the "slack" side of the chain so is never under significant load.

The only load is that required to pull the chain back from the crankset under enough tension to stop it from drooping, which is provided by the spring in the derailleur cage.

All the load in the chain system is between the cassette and the chainrings on the top.

Perhaps a link to the article will reveal more? There's a lot of BS in bike technology, so I'm automatically sceptical!

Maybe load was not the appropriate word to use.

Smith explains that that friction in a chain-based drivetrain is created largely at the eights points of articulation, where the chain bends around the chain ring, cassette and pulleys.

"Any time a chain articulates, friction is created. And any time it disengages, friction is created," Smith said. "When you think about pedaling 95rpm, you are looking at 40,000 stiction points a minute."

In the DrivEn system, those eight points are replaced by four points, each of which rotate on ceramic bearings. The chain ring's teeth and cassette's cog engage with the bearings on the shaft, which itself spins on bearings.

https://www.bikeradar.com/road/news/article/ceramicspeed-dri...

98-99% iirc for a fixed (track) gear setup.

There has also been a recent trend on the track towards larger chainrings and sprockets for a slight increase in efficiency, as the chain has to go through less of a tight radius, whilst still maintaining the same gear ratio.

I don't think the big cog will be stiff enough to not bend under the pressure.

"while the 13-speed rear cog looks like the unholy union of a compact disc and the Sarlacc pit from Return of the Jedi"

After looking at the thing, it is a remarkably apt analogy, too. I wonder how long it takes to machine.

Does anyone know if this would this be UCI legal?

It's just a concept. CeramicSpeed have said they would like to partner with someone to design the shifting mechanism[1].

[1] https://cyclingtips.com/2018/07/ceramicspeed-driven-drivetra...

Not a chance. That gear will rip your leg to shreds in an accident.

Cassettes with similar gear ratios seem just about as leg shredding. Then again anything at 20mph eats legs. I'd rather put my foot or hand into the article's gear, then into the spinning spokes of a wheel.

Look closely at the gear: the teeth are standing out on the side at right angles to the plane of rotation and are not protected by the geometry of the frame in any way.

That means that they will be very efficient at shredding flesh, far more so than a normal cassette which has the fastest rotating surface closest to the wheel.

Obviously you don't want to be in contact with either if you can avoid it.

As for the spokes: that's a problem that actually reduces above a certain speed because the spokes will start to act as a plane and throw you out before you can get anything properly wedged. At intermediary speeds they are really dangerous.

That's why it is important (especially for kids) to wear proper shoes on bikes and to put spoke guards on the rear wheel where feasible.

It doesn't take much to have serious injury of toes or heel, including damage to the Achilles tendon.

that gear is perfectly flat-ish and could accommodate a fairing quite easily.

Yes, it could (with a little protrusion for the drive gear).

At least on the other side it naturally sits with the flat side out.

I have a large scar on my calf from a conventional chainring. If your chain is on the inner ring, your crankset is basically a blunt circular saw.

Ouch. I very narrowly escaped that (but did break my leg :( ). On intermediary 'sports' bikes and mountain there typically is a guard ring mounted on the outer gear ring to mitigate some of that risk. Most road bikes don't have them though.

Would this design not be... less of a chipper shredder if they switched the cylinders and the teeth?

Yeah, I could imagine that getting the end of your jeans or a shoelace caught in that, you're in for a bad time.

Generally if you have shifting, you have a freewheel. If something gets stuck, no big deal, just stop pedaling.

heard of covers?

I suppose you could do a number on your hems, but I’m more worried about feet and calves. I’m thinking about a crash situation specifically. That thing is more of a meat grinder than cogs. Granted, the front chainring is more of a meat cleaver on a traditional bike. I knew a guy who got a nice free tattoo on his right calf. Still gives me the heebies...

Reply on news.ycombinator.com