Ceramic-bearing bicycle drive shaft (2018)
bikeradar.com
bikeradar.com
They put it out as a marketing thing. It had a bunch of impractical caveats.
They haven't announced they solved any of the problems with it in the year since they put it out.
The friction thing & drivetrain is mostly a non-issue. If you're riding fast enough to care the % drag from the chain/drivetrain gets smaller and smaller the faster you go. It's pretty much a non-factor compared to air resistance.
You get way more benefit from doing some extra stretching and yoga so you can lower your handlebars and still ride hard than you do changing to ceramic bearings or fiddling with expensive lubricants or chains/belts/shaft drives.
Compare to 10mph (average speed for most casuals) where your total wind + drivetrain is about 50w of energy. Wind resistance goes up by the cube, drivetrain is mostly linear.
What do you consider a steep hill? There are hills here that I can routinely get up to 35 mph with a little effort. Many cyclists have gone downhill at speeds exceeding 50 mph [1].
Or they should. 30 MPH is fast on a bike. If anything goes wrong, it's going to hurt.
The only thing of going wrong i can think of is insects in the eye, and generally unpleasant airstream in the eyes. Use sports glasses. Problem solved. Anything else is situtational awareness, maybe a helmet, and good maintenance.
Down a steep hill on my road cycle I can easily get to 45-50mph.
Rolling resistance doesn't change that much with speed[1].
Rolling resistance energy balance at 18mph is on the order of 10-15W[2]
Drag increases with the square of velocity (D=0.5 * <drag coefficient> * <representative cross sectional area> * <air density> * <velocity>^2).
The drag energy balance (plus some rolling resistance) at 10, 20, and 30mph is on the order of 10, 100, and 300W, respectively[3].
[1]https://www.schwalbe.com/en/rollwiderstand.html (see section "what exactly is rolling resistance")
[2]https://www.bicyclerollingresistance.com/road-bike-reviews
[3]https://www.exploratorium.edu/cycling/aerodynamics1.html (see the calculator)
30mph does not require anywhere near 1000w, it's more like 350-400w on flat ground depending on the riders position and size. I'm a former amateur racer. I've done 30mph steady state for a mile on flat ground without a draft before. That's a 2 minute mile and there is no way on earth I could ever do 2 minutes @ 1000w. 1000w is more like a 5s sprint for an amateur, you would put that power out in a burst to accelerate and probably end up doing about 34-35mph on flat assuming no draft. Pros are probably more like 1500w. 30mph is nothing. I probably exceed 30mph every time I get on my bicycle if I'm riding for exercise as opposed to commuting. Almost no race goes on where speeds don't hit 40mph, and 50-60mph is not rare.
Rolling resistance is not drivetrain resistance. Most of rolling resistance is tire resistance. It doesn't scale up at the same rate as aerodynamic drag.
Chain resistance is much smaller than tire resistance. It does go up with power but not really with speed directly because you shift.
50 km/s is over mach 145 :)
Likewise, poor cooling can lead to a 10x decrease in motor performance.
Cooling is usually the biggest limiter of power, but max rpm (before the thing blows apart) combined with iron saturation (effectively meaning the magnetic fields inside can't get any stronger, even with more current) can also be a limit.
Friction Facts started out as a guy (with an engineering degree) that developed his own test equipment to measure drivetrain losses under load. He eventually published his own wax formulation so that a DIYer could replicate his best effort at reducing chain friction. It produced a chain that had about 4 watts of loss, which is pretty good. That's over 98% efficient for a person putting out 250 watts. 250 watts is high for someone that's never ridden a bike, doable for someone that commutes by bike every day, and a pro wouldn't even break a sweat. So going to 99% improvement, or "reducing friction by 49%" sounds like a lot, but you're only gaining back 2 watts out of 250. Also, Friction Facts felt really cool when is was just a backyard engineer publishing his results, but then he got bought and took all his results/website offline. Luckily the internet archive and many forum posts have the major results, but it felt kind of like he was selling out to the man. I understand that he has bills to pay so I'm not mad at him. I also understand that a regularly lubed bike chain has higher than 4 watts of loss and most people aren't willing to melt wax on their stove to wax their own chains (I do though).
The CeramicSpeed concept couldn't shift in 2018 and they haven't demoed a shifting version as of August 2019. The pie plate at the back is only for looks, and claiming 13-speeds is totally worthless at this point. Heck, I could CNC a 30-speed cog for the back, since it would fit inside the diameter of the wheel. I predict that when a shifting system is functional, it will increase the complexity and add to the friction of the drivetrain, bringing to about the same level of friction as top end chain based systems from Shimano, SRAM, and Campagnolo.
This is a promising idea, but ultimately it's worth more as a way to avoid patent infringement than it is to make a faster bike. Another smaller player, Rotor, makes a 13-speed system that shifts hydraulically. Rotor's marketing claims that hydraulics shift better, but everyone knows that it's just a way to keep from stepping on the established players' toes.
Tire rolling resistance costs about 40 watts (on regular roads with road bike tires; less in a velodrome and more with knobby, thick mountain bike tires).
The bike itself costs 8 ~60 watts or so in air resistance, less for aero, deep-dish tri bikes and more for round-tube wide-tire gravel bikes. But more importantly, air resistance increases with the square of velocity, no matter your starting drag coefficient.
The remaining 150 watts or so are caused by rider air resistance.
And of course, all 250 watts are produced by the rider and dependent on their fitness level.
Clearly, the best way to go faster is to put more energy in. Next, improve your posture to produce less drag. However, it's easier to spend money than to sweat and suffer, so people instead turn to aero bikes, tires, and the drivetrain.
not disagreeing with anything you're saying, however at the super-elite level everyone puts in the maximum amount of perfectly allocated effort into training and preparation. if we could eliminate 4 watts of drive train friction loss that would be material. For your average weekend warrior the points you make are definitely true, but hey, not the worst thing to blow your money on...
Can't imagine this catching on. Ordinary chains are already really efficient, and the efficiency of this will only last until dirt gets into all those tiny little open bearings. (Sealed bearings would eat into the efficiency gains.)
I believe it uses the Bionx system, which uses a direct hub drive. The company is lo longer in business, IIRC.
Not necessarily any battery, but the electric cabling replacing chains.
Jeasus guys - engineering can solve ALL of these issues.
My understanding is that the tradeoffs are:
1. Weight
2. Cost
3. Lack of user serviceability.
It definitely seems like a good tradeoff in some circumstances - particularly in rainy environments.
My understanding is that you can't maintain the hub shifter at all. It's sealed so it's generally pretty low maintenance, but if/when it eventually breaks you have to send it to the factory for repair.
The internal gear hubs have been perfected before derailers even became a thing; it's a very mature tech.
You can[1] do some maintenance on them. I don't have any experience with that, as all my internally-hubbed bikes were stolen long before such a point :-) :-(
One big tradeoff you missed is efficiency. The belt drive is efficient, but internally geared hubs are currently less efficient than traditional derailleur setups. Losing watts to drivetrain loss is a major reason that internal hubs are not as popular with human powered bikes. The durability and maintenance need to be WAY better to get someone to switch over.
I'm surprised there are retrofit kits. With a belt drive, usually one of the chain stays has a gap in it so you can get the belt on and off (you accomplish the same with a chain by taking a link out to install/remove the chain, but this is not an option with a belt).
See page 63 for efficiency and 64 for longevity.
The graph on page 64 is confusingly labeled, but the numbers indicate "watts expended before wet dirty run" - "watts expended after".
But it's 3-5x the price of chain equivalent setup.
Not at all: the chain goes through a closed triangle in the frame by connecting two loose ends. The belt comes as a closed loop and needs a frame that can be opened. Additionally, frames build for chain gearing (instead of internal gears or single-speed/fixed) lack tension adjustment by either a horizontally adjustable dropout or a bottom bracket excenter mechanism.
Sliding dropouts are undoubtedly preferable, but any modern frame with a BB30 or PF30 bottom bracket can be trivially converted to an eccentric bottom bracket.
And of course the loss of efficiency due to belt drive is also pretty bad.
I am convinced that this will be my last belt drive bike.
Here‘s a (German) Video from a guy with a severely damaged (but still rideable) belt after 4500 km:
https://m.youtube.com/watch?v=n16gN7RDWEY
In the comments you‘ll find people whose belts snapped at 1050 km, at 1000 km and one at less than 2000 km.
I don‘t know how likely this is, it might be quite rare. But since it‘s non-repairable it‘d make me uneasy on longer bike rides.
Belt damage? They put these belts on motorcycles, it's possible to damage them but they are insanely durable, probably stronger than your typical chain. Somewhere I picked up a staple on mine, I'm going to get it replaced but I've done several hundred miles on it with no issues.
The IGHs are a different matter, they're pretty sensitive and if you get them off a little bit they work but your shifting will be hosed up. Bike shop fixed it all though. My advice? Don't mess with it, get it set up, make sure the little connector on the cable is tightly attached and you probably won't have issues, I had some slip on mine.
I've been running Conti touring tires with some mixed tread and pre-slimed tubes, I've not had flats but I've worn out 3 or 4 sets of tires. It's not a big deal to change it though. It's a commuter bike so it has bolts rather than quick release skewers, I'd use slime and maybe even kevlar liners if you get a lot of flats and just make tire changes a wear issue. It's a heavy bike, it's not a carbon roady by any stretch but it has been as close to zero maintenance as any bike I've had. It's actually required more maintenance on the disc brakes than the drive train, they were getting a little squishy but just rebleed them and they're good again. The only "issue" I've had was when I got a lot of dust on the belt, it was kind of squeaky for a while.
This carbon/ceramic drive shaft is kind of interesting. I always thought that the chain to pully and gear touch point was where the friction loss was, not interlink on the chain itself. For years and years, they've offered bigger pullies for reduced friction. This drive shaft will still have a similar touch point; also seeing as how both pieces rotate, it looks like there could be side to side friction as the cog on the shaft touches the gear on the wheel and they both rotate unless the cog and gear have some very special shaping that I'm not seeing. As the cog rotates, when it first can engage the gear, just the tip will be touching the gear but as it continues it will push more of the tooth on the cog in to the gear, am I wrong on this?
For what it is, I love it, it's not a racing bike though. I'm curious about the mtb uses with a Rohloff hub, it's a fair amount of weight.
[1] https://www.hebie.de/en/protection/chainguard/chainglider/35... ?
Are bicyles a fashion statement, or what?
"Seht her wie geil mein Riemen ist!"
Mine also has the Alfine 8 speed hub, which is kind of suck, but I make it work. I recently had it re-lubed (dunked) at my LBS, which seemed to help. I just wish there were some nice IGH in between the Alfine 8/11 and a Rohloff. I'd gladly pay for something better than the Shimano.
For commuters, I cannot fathom why people would deal with a chain when belts are a thing. Plus, good chains aren't cheap (I like Wippermann) and when you rack up miles you need to replace them enough that your wallet notices.
Tires: +1 for Schwalbe Marathons. Four years and no flats. NB: if you are worried about weight, upgrade to the Marathon Supremes with the aramid (foldable) bead. ~$75USD/each but totally worth it.
Oh yeah: a dynamo hub is a life-changer. My Focus came with a SP PD8 and lights, stock. Never having to strap on a light, or remember to charge it is a revelation.
Ride safe!
However, my bike is my car. I'd no sooner swap out my own bike tire than I'd swap out my own car tire. Not that I can't, it's just not something I enjoy doing. My hub-geared, belt drive, hydraulic-disc bike is incredibly difficult to maintain... but I also just bring it in to my local bike shop once per winter. That one trip is all the maintenance it needs for the 3200 miles I put on it per year, mostly in the rain. I'd never go back to lubing a chain every week, putting new pads on every month, replacing the chain every 3, etc. 6 months after I bought this bike I gave away my repair stand and it was glorious.
In my neighborhood I get a flat every other month. If I had to pay someone else to replace inner tubes it'd cost as much as driving a car.
They can be messy, but I've lost count of the number of self-healed punctures I've had over the years. A small loss of pressure and off you go.
Currently running WTB Venture 47c [changed from Horizons] on a 650b 'gravel bike', fantastic stuff, gets me where I wanna go.
If your bike is your main form of transport and you can’t change tubes, you need to have two bikes. Unless you live next to a 24 hour bike shop...
Oh no not at all! I hate swapping tires. I live in an expensive and cold country. Swapping bike tires costs $50. I couldn’t afford that twice a year (because of winter tires) I have to do it out of desperation which is why it’s more work than the maintenance benefit I get.
I bought a Ghost brand bike at REI (they're a German brand, but REI is the exclusive US distributor). The model is an "Urban" of some sort, can't remember which. It has an 8 speed Shimano Alfine internally geared hub. I really wanted a Spot Acme (made in Colorado) that was absolutely gorgeous, 11 speed hub, Gates Carbon drive, brooks leather saddle, etc. But that $2k+, and the Ghost was on sale from 1300ish to $850 if I recall correctly. It didn't have a dyanamo so I bought a Shutter Precision hub and had a local bike shop build up a wheel with a dynamo so I could run lights.
The Ghost came with a Continental belt drive instead of the Gates. Instead of having the center groove of the Gates, the continental just... stays in place. Hopefully. The right side of the crank had a flange, and the left side of the hub gear had a flange, but the belt could technically fall off the 'other' side of either one. And it did, occasionally.
Belt tension is regulated by how "tight" you make it by adjusting the rear wheel in place. The Ghost frame has angled dropouts sorta like a normal road bike (not horizontal dropouts like a motorcycle or like the Spot bikes I saw). It's a COLOSSAL pain in the ass for the average person to hold a heavy geared hub wheel in place with one hand, thread the nuts on with your second hand, use a wrench to tighten the nuts with your third hand, and deal with belt tension with your 4th hand. Not to mention keeping the wheel straight. And if you don't get the tension right (how many lbs of tension for a 1 inch drop, again?) the belt will fall off. Repeatedly.
Another problem is that you need to unhook the shifter to remove the wheel. It's not that hard, but when you're a novice to these bikes, it's tricky because some models recommend you shift to the lowest gear; others to the highest. You need some needle-nosed pliers and some finagling to get the cable out.
A problem specific to the Ghost was that the rear disc brake rotor ran into the caliper as you tried to slide the wheel out. So you ALSO have to unbolt the caliper to remove the rear wheel. Overall, it's a vastly bigger pain in the ass to remove the wheel on that bike than on a "normal" bike. God help you if you have a flat on the side of the road. I went through a bunch of different tube/tire combinations before I ended up with slime tubes, which i now HATE because the valve stem gets gummed up and leaks constantly, but I've spent a year or more dealing with it (adding air every week or so) because I'm too lazy to go through the hassle of removing the rear wheel to take out those godawful leaky slime tubes. (but at least I found a tire that's great and puncture-proof!)
Anyway, back to the drive belt. After owning the bike for roughly 5 months and 1000 miles, while trying to cross a busy street, the belt snapped. I tried to source a replacement, but couldn't. I took the biek back to REI, where it sat for the better part of a month while they tried to find parts.
To their credit, Ghost had discontinued use of the Continental system in favor of Gates, and as a goodwill gesture, they replaced the entire drive system of my bike with the Gates drive system.
Knock on wood, the Gates system has been perfect, doesn't squeak like the continental system does, apparently isn't as picky about belt tension, won't slide off (because of the center groove).
The shifting is less than perfect but I like the fact that geared hubs can shift in reverse, so you can downshift at a stoplight while waiting for a green. The spacing between the gears is a bit uneven, but I think it's a pretty good low-maintenance solution for a commuter bike.
But it's not perfect.
You mean the internal gear hubs?
They've been competitive for over a hundred years[1].
I had a bike with an 8-speed SRAM[2], it worked fine (got stolen one day).
But has some tradeoffs that mean it isn't quite there yet.
I'd have guessed it's more like 99%. isn't anything that isn't heat or noise = perfect efficiency? a normal bike chain doesn't get hot or make much noise at all, does it? seems to me just about 100% efficient. maybe transfers a bit of vibration to the rest of the frame? but not 8%-4% of its transferred force....
Motorcycles can have 150hp engines (~100kw) and even then the chain doesn't get particularly hot.
This is doing well as a fake-halo product for Ceramic Speed and their existing products, it is a crowded marketplace and not many people care about bike bearings, this product makes people aware that there is room for efficiency gains.
Actually, it's not - take a look at the renderings again.
If you notice (and the article details) the rear "cassette" is actually a flat disk with concentric "gearing" which the roller bearing "gear" on the shaft interfaces with.
Switching gears would be accomplished (in theory) by moving the roller bearing "gear" in and out along the radius of the rear disc "gear".
It reminds me in a way of old wood "peg" and "lantern" gearing seen in wind/water mills and similar old power transmission systems, except in this case the "peg" gear is a formed or cast disc, and the "lantern" (which you couldn't shift due to the plates at the ends) is a "ring" of bearings, the bearings forming the "teeth" of the gear (which isn't that strange, if you understand how a simple gear is basically formed out of partial circles along the edge of a larger circle, all "joined up").
They do mention though, at the end of the article, that they can't shift gears easily and something about an internal wireless servo (which I think is kinda daft). I would instead use a splined shaft, and just change the "length" of the shaft to move the end-gear in/out of the wheel gear, and actuate it with a normal lever and bowden-cable type arrangement, though that might interfere with their efforts/want for more efficiency...
There is this difference between theory and practice. So even if there was a solution - yours is elegant b.t.w. - then it would take a whole lot of testing and fine tuning to get right.
I am actually quite astonished at how bicycle parts evolve, the derailleur took the best part of a century to get indexed gears and those took decades to go from clunky to smooth. What took so long to get right? Scores of refinements were needed, e.g. to the sprocket teeth profiles, to being able to manufacture a decent chain to basics of friction in cables. Then there is the challenge of mass manufacturing these things so they are durable and precise.
I am not dismissing the ceramic speed shaft innovation - it gets people thinking - but I doubt my bike will be losing its chain any time soon!
It's a great idea, and I do love innovation. I've honestly never really had a problem with a chain though, myself.
I have about 6,000 miles on my cross bike (that gets used in some pretty dirty, dusty, sometimes-muddy conditions). And maybe 2400 on my newer road bike. I just clean the chain with wd-40 when it's dirty (sometimes water/soap first if really muddy), dry it well with a rag, then apply some liquid chain lubricant (sometimes after every ride, if it's needed).
"[...] the company had already pushed the envelope with lubrication, bearings and pulleys about as far as possible — creating a roughly 98% efficient driverain"
“The quest was, how can we get a 99% efficient drivetrain?”
But you know what might be an interesting design idea:
Make a shaft drive like this which is centrally aligned. But the bike has TWO rear wheels as closely spaced as possible and the shaft drives a single axle for the rear -- but has more gearing up front...
But make the two rear wheels with a camber such that they angle out slightly such that their centerlines are as clos as possible - allowing for turning to be essentially the same as a single-rear-wheel bike...
Now - there might be a way around this, but it would probably introduce more complications and friction to the mix (one way would be to drop/raise the central shaft off-center of the driving wheels centerline, then have separate fore/aft gearing on the shaft to drive the wheels in the proper direction).
Yeah... basically what you said in the second paragraph of your comment....
This would be a fun design challenge. Anyone Solid on Solid Works want to give it a go???
---
OH . I thin I have it, but I dont know how efficient it would be:
A Worm Gear:
A differential gear solves both the direction problem and the distance between the wheels when turning problem.
Instead, though - flex might be able to be eliminated by making the "hub" larger, so that the disc (according to the article and renderings, it's a flat and not conical system) has something "solid" on the backside (this isn't shown in the renders from what I could see).
I'm not sure what such a change would do to the performance or handling of the bike, due to the shorter lengths of the spokes to the rim, and the larger mass of the hub, and who knows what else. I haven't been on a bicycle in well over a decade, and even then it was just a cheap Huffy I bought at Walmart for Burning Man use...
So one could say that this technology, even if it were real, would not be targeted at me as a consumer or bike user...
The replacement rate for those bearings would be astonishing.
However, as long as there are people with $250 to spend someone will make a $250 set of bearings- don't forget to lube them with snake oil for the fastest results.
"Hybrid ceramic bearings are the equivalent of trying to run a locomotive on an asphalt road - the hardness differential causes the road (raceway) to become damaged."
[1] https://www.hambini.com/blog/post/ceramic-bearings-vs-steel-...
Your best value if you want fast bearings is to clean out the grease and replace it with something lighter like sewing machine oil or Bones speed cream. The downside is that you have to maintain it more often because oil doesn't stick around as long as grease does.
Also, the difference between ceramic and regular bearings is on the order of 1 watt. You'd save way more time in a BMX race by wearing tighter fitting clothes, and if you are doing tricks instead of racing then ceramic bearings on a BMX bike makes no sense.
This basic idea has had over a hundred years to take off.
The particular design seen here appears particularly impractical.
The rear shifting mechanism has a poor selection of gear ratios for road use. The gear ratios are not sufficiently close. In my bike I have a Shimano Claris (or Sora?) cassette like this: with 12-13-14-15-17-19-21-23 teeth:
https://i.imgur.com/BFFBYzj.jpg
Even if the drive train has a bit more efficiency, if the gear ratios are bad, that will be a deal breaker, because bad gear ratios wreck the efficiency of the human power plant.
The system has to be extremely rigid for the gears not to skip. If there is any flexibility in the system, it requires massive preload to hold the parts in contact. The chain solves this problem; its own tension binds it to the sprocket teeth.
The rear engagement mechanism has a kind of gear made up of a dozen tiny bearings. Once dirt from road spray gets inside those, they are done.
This is a basically a fair weather toy for yuppies who don't know anything about bicycles.
If I understand it correctly, the issue is that the driveshaft has to move from one "sprocket" to the next in a very small amount of time at high rotational speeds of the cassette. Because of the weight of the bearings at the end of the driveshaft, a very large force is required to accelarate it.
Before ball and roller bearings, though, it was really bad.[1]
What a funny coincidence
(Not implying anything about the company in article, the article is from july 2018)
If you check tomorrow, it’ll say “2019-08-17 Application status is Expired” instead.
A conventional chain in good condition is >90% efficient. How do you double that?
I need one of these YESTERDAY.
How get.
Ill trial the fuck out of this thing.