Hacker News new | past | comments | ask | show | jobs | submit login
Chainless electric drive system “Free Drive” for bicycles (2021) (schaeffler.de)
104 points by Tomte 15 days ago | hide | past | favorite | 253 comments

I wonder how good they can make the feel without a direct connection to the road?

For example, as you stand on a bike stopped at an intersection you have your foot resting on the petal, ready to take off. As you launch, you put a lot of force on that petal and rely on the feeling of connection to the ground to get going.

This motor is going to have to hold that force statically and have a control system with sufficient power and bandwidth to emulate the familiar feeling of the ground. Not impossible I guess, but I wonder how well it works.

Regarding efficiency, I think this is a smaller issue than it seems. Let's say a chain is 95% efficient and their system is 85% efficient (they claim 5% less efficient than a chain, apparently, but I'm sure that's a stretch). Most energy cycling, if not grinding up a steep hill, is dumped into air resistance. But speed only goes up with the cube root of power with regard to air resistance. This means (given the hypothetical numbers above) that you'd go 3.8% faster with a physical chain (or 1.7% if you believe their loss numbers). That difference is not perceptible on a bike and not an issue for 'getting around' use cases.

It's interesting that after 140 years, the behavior of a chain drive safety bicycle with pneumatic tires is still the gold standard. Virtually everything that's noticeably better on my bike is related to the materials used. Although, variable speed gearing was a nice invention, but still, 100+ years old.

Indeed, the chain+derailleur system is incredibly lightweight, efficient, simple, and also fairly reliable. It's one of the most elegant solutions in mechanical engineering, imo.

I disagree, I find chain and derailleurs to be annoying and unreliable in the worst way. I'll grant them their crown of efficiency, but I hope to see more refined CVTs. I currently have a bike with a NuVinci and I strongly prefer the way it functions over the clunky must-be-pedalling shifts of a derailleur. As an added benefit the system is contained, presumably hardening it against damage and reducing maintenance.

How do you find it unreliable? I put a couple thousand miles on mine a year and it sits outside. Occasionally it gets hit with wd40 then lube and a rag. That's all I do with this thing. I don't get all the complaints from people on the fiddling with derailleurs. I've never had to fiddle despite years of riding. Just some shimano components mostly 105 and ultegra but nothing special or too new. I'm still riding 2x8 actually. Maybe the fewer rear gears leads to fewer issues with alignment? I do find that I can shift into any of the rear cogs easily with either front cogs, very permissive groupset to being in the wrong gear at the wrong time too and sloppily finding another one (a lot of chaos sharing the road). Maybe I'm just not attuned enough to feel when the gearing is going out though.

They are reliable enough but not exactly perfect. It’s pretty common to have the chain slip off. Even on a well adjusted setup the chain can still slip sometimes. It’s just acceptable because it’s easy to pull back on. But if it was a deeply internal part like on a car, it wouldn’t be acceptable

> They are reliable enough but not exactly perfect. It’s pretty common to have the chain slip off.

It should not be common for a chain to slip off, something is wrong. The something may just be parts quality.

Derailers are a case of getting what you pay for. The cheapest ones are junk and will never shift well. My partner bought a bottom quality bike with no-name shifters and it has never shifted correctly from day one. It does also drop the chain. No amount of aligning will get it to shift well.

In contrast my 18 years old mountain bike with Shimano XTR derailers has always shifted perfectly and still does.

Bike derailers is one area where it's definitely worth it to go for top of the line components and never worry about it again. Low quality components will make you hate they money you saved every single time you attempt to shift.

Maybe I've just had bad luck, but two of the bikes I've had have come apart. The derailleur gets misaligned for whatever reason. One of them I had to take to a shop because despite extensive experiment I could not get it to properly align, and I don't know that they did either. The other was perhaps just old - an '80's era steel frame road bike from a Sears catalogue I donated to a repair shop, and it eventually got exploded for parts. I suspect the third would've succumbed to the same fate. It may just be me mishandling them, but if they can't handle this style they're not fit to ride. :cool:

Also I hate dropping gears in the middle of an intersection.

They got a lot better if you were buying quality bikes (not department store bikes) in the 2000s. I have a road bike that has 20k miles that never needed shifting adjustments besides replacing worn out chainrings and chains.

I busted a Kona Splice's drive train, started ghost shifting and I couldn't figure out how to get it adjusted. I had a Cannondale that I never had problems with but it didn't get used a lot.

Maybe it's hopping curbs? Can't be helped here, the roads just aren't designed for bike commuting and everybody has a lifted 3/4 ton they can't fineagle into their garage. It's lovely.

"Maybe it's hopping curbs?"

Can't see how that would affect the shifting, but if you are graceless you can beat up the wheels enough to cause them to get out of true. If you have my wheelsets, 32 spoke 3x on HED Belgium Plus rims that I built myself, you can hop curbs all you like. I do, and I ride non-technical dirt too.

As for shifting, I've been riding for 45 years and have never had a problem with manual shifting, from the dark ages 5x2 speed to my current 9x3, 10x2 and 11x2 rides. Frankly I am happily shocked at the precision and durability of my newest bike, the 11x2, given the narrowness of the chain. After 2.5 years of 3x weekly 3000' climbing, including some dirt, riding through the winter, and the drive train is dirty, no missed shifts and no noise. Amazing. I would never have guessed that.

I think the key is matching up the right parts. I ride with a 1x11 mid grade set up that was made for each other and after dialing it in I have no issues either.

Today's mid and even sub-mid tier Shimano drivetrains are an amazing combination of performance, reliability and price. 4-piston disc brakes are also fantastic and suspension forks are great too. There's a lot of BS in bike manufacturing (ex: new standards) and carbon frames need a serious rethink but bikes today rock!

I have 30 year old derailleurs with thousands of miles on them that are still going strong. I don't really understand the reliability complaint here.

I find the chain+derailleur system to be less reliable than drive-shafts, but it's trivially repairable for a 10 year old with no tools, which is a huge win.

I have left this behind for a Rohloff Speedhub 500/14 and I'm not looking back. The chain runs inside a Hebie Chainglider protected from the elements and is lubricated with a nice mineral oil. I can run this setup without any maintenance whatsoever for 2000km in every weather. The Rohloff itself will likely outlive me, provided it gets its oil change every 5000 km or so.

I do utility cycling and long distance tour cycling.

Ehh... not entirely true. It's the "gold standard" because of cost and (lack of) need. It's an elegant solution for non-functional cycling.

Band-drive systems with internal hubs are superior for commuters who need reliability over anything else.

Tubeless tires are now the standard for riding on irregular terrain.

Disc-brakes are not the standard for hilly areas or people who ride at high speeds.

Meanwhile, single chain-drive fixed/free gear bikes are still perfectly serviceable for most use cases (even if you need to select from a few different cog sizes to fit your area).

The reason why "chain drive and pneumatic tires are the gold standard" is that most people don't ride bikes. If people did, you'd see much, much more variation in technology on the street, as they would be used for functional purposes, rather than recreational purposes.

Some places have lots of functional riders, such as in northern Europe, and AFAIK they mostly use the gold standard.

It can work both ways. If there are already lots of functional riders, they they probably have lots of functional bikes which means: They're confident about getting reliable use out of existing technology and are less likely to see new components as a silver bullet. In my case, none of the new developments are exciting enough to motivate me to replace my bikes, which work just fine.

Off road cycling is its own beast, with the development of specialized bikes for that use, and also the design of trails that challenge both your skill and your technology. But the old footpath through the woods, or gravel road, is still the same as it ever was, and a regular bike handles it just fine.

In the US, there's a perpetual effort by the bike companies to get people out riding, when everybody's already got a nearly brand new bike hanging in the garage. They want to sell new bikes, so of course the silver bullet is new technology. I think the rapid adoption of e-bikes shows that the real barrier was not the detailed performance of particular components, but physical effort. Who knew?

> I think the rapid adoption of e-bikes shows that the real barrier was not the detailed performance of particular components, but physical effort. Who knew?

In a lot of cases I think it might be less about effort itself, and more about climate and what the level of effort means in that context. I know where I live half the year the average temperature is 80F or above, and pedaling or not could very well be the difference of showing up as a sweaty mess that needs a shower on site to appear presentable or not.

> I think the rapid adoption of e-bikes shows that the real barrier was not the detailed performance of particular components, but physical effort. Who knew?

Is there data on this rapid adoption? I'm not seeing it. It's hard to believe riding a bike is so much effort, but who knows?

Belt drives are fine, but I don't think they're really the improvement over chain drive they're touted as.[0] If you really want reliability out of your bike, they're both sensitive to dirt and grime and benefit from a full chain case.[1] Definitely a fan of internal hubs for commuting, but they're usually at least an order of magnitude more expensive than derailler systems.

Both enclosed chain cases and internal hub gears have been around over 100yrs though, almost from the beginning of bicycles.

Surprised it took so long for tubeless tires and disc brakes to become popular on bicycles since they've been used on motorcycles and cars for many decades. Otherwise the "gold standard" fancy bikes still exist and still look closer to a bike from 1900 than most bikes on shelf do.

[0]https://www.sheldonbrown.com/gloss_ba-n.html#belt [1]https://www.sheldonbrown.com/gloss_g.html#gearcase

Same thing with spoked wheels. The same design we use today would instantly be recognizable to someone from 150 years ago. (e.g. https://www.thehistorypress.co.uk/articles/from-discomfort-t...)

I imagine spoked wheels would be recognizable by ancient chariot riders even.

They wouldn't recognize them if they looked closely. Chariots, I'm guessing, had radial spoked wheels - the spokes are radiuses from the hub to the rim. Those were tried in early bicycles but didn't stand up well to usage. That led to two innovations in the late 1860s - mid-1870s that persist today:

* Tension method of spoking: Invented by Eugene Meyer in Paris in 1869.

* Tangential spokes: If you look closely at bicycles, the spokes are tangential to the hub, not radial. The tangent-tension combination (and maybe tangential spokes independently) were patented by James Starley in 1874. The "spokes are placed so as to be tangential to the hub in both the forward and the backward direction, thus forming a series of triangles that brace the wheel against torque during either acceleration or braking." [0]

The improved spokes were introduced before chains and gears, allowing larger wheels. Larger wheels served as a subsitute for gearing, increasing the ratio between linear distance traveled and one rotation of a pedal, and thus the big-wheeled 'penny-farthing' or 'ordinary' bicycles were born.

However, knowing the history, I haven't yet grasped the mechanics of how tension and tangents benefitted wheels, beyond general concepts. Does anyone know a good technical source (not Wikipedia)?

[0] "Bicycle Technology" by SS Wilson (Stewart Wilson, afaik) in Scientific American (March 1973)

Im pretty sure if they looked closely they would just learn a new and probably better way of doing something they already do. They would see the wheel there.

True. One difference is that bicycle week spokes are tension-based, whereas I assume chariot wheel spokes were mostly about resisting compression.

I don't think they are going for road feel:

The central component of the Free Drive system is the Schaeffler generator, which sets the constant resistance on the pedal while simultaneously absorbing the rider's pedaling power

So I think with this bike when you take off from a stop, you just squeeze the throttle or whatever controller it has and take off using battery power.

On the other hand, you can pedal at your most efficient cadence/force all day long without regard to terrain.

> So I think with this bike when you take off from a stop, you just squeeze the throttle or whatever controller it has and take off using battery power.

As they are German, I think that's not the case (if they want to be classified as an eBike). To be classified as an eBike it needs to work in "pedal assistance" mode. That is, you can't have a throttle/button/whatever, the engine only starts if you are pedaling.

25kph eBikes per EU regulations are allowed a throttle if it cuts out at 6kph, IIRC. Just for starting, in other words.

On my e-bike it's more of a walk assist tool. Press a button and the bike will push itself uphill. It's nice, considering the weight of the beast :)

i doubt it's designed for classic single-person e-bikes; there's one picture at the bottom of a cargo-bike with a person-sized transport box. i think this system is for small inner-city utility vehicles for delivery services and craftsmen (transport of a limited set of tools and material).

the low overhead of bikes (registration etc.) isn't necessarily a problem for a company's motor pool. iirc it might very well be possible to ride those on bike paths - even if they need a moped registration plate - as long as the maximum power and speed is limited.

in austria, small electric numberplate-less mopeds (scooters) with a 25km/h speed limit and no pedal assist are legal, no idea about germany tho.

Ok, then you start pedaling, but since they specifically say it provides constant resistance, it's safe to say that they are not emulating road feel.

Having kick-started a bike uphill from many stop-signs, let me say that road feel is 100% overrated

European ebikes are also limited to a 250w motor. They must need much more to make this system work, so probably can’t be classified that way.

Most "average" riders don't put out more than 100 or 200W when pedalling. I'm pretty sure 250W is plenty to get a bike going.

>Regarding efficiency, I think this is a smaller issue than it seems. Let's say a chain is 95% efficient and their system is 85% efficient (they claim 5% less efficient than a chain, apparently, but I'm sure that's a stretch).

Would be interesting to see proof of their efficiency claims, having looked into this before series hybrid drivetrains are in the 80% efficiency range, you have generator->charge controller->motor controller->motor losses.

There are no production series hybrids I am of aware of in automobiles, all hybrid seem to have a parallel component connecting engine to wheels at highways speeds because its much more efficient.

Trains and large boats do diesel electric drivetrains but its not for running efficiency but other factors like traction /throttle control and power routing.

Its very difficult to out perform a mechanical drivetrain in both weight and efficiency if your power source is mechanical (ICE / human body).

Trains do it because they need to apply a lot of torque at zero speed to get moving. The only way to do that with a mechanical connection to a combustion engine is with a clutch, which would burn up from friction before the train got going.

In theory a hydraulic pump would work, but I'm sure that electric motors are better for that usage, since that is what they use.

There's also torque converters, which are a hydraulic mechanism. This is the basis of most (non sporting) automatic transmissions. Higher performance torque converters have a lock up mechanism that functions as a clutch making it rigid at high speeds. So you can kinda get both.

> There's also torque converters, which are a hydraulic mechanism. This is the basis of most (non sporting) automatic transmissions.

By virtue of being the dominant automatic transmission in the US, where automatic transmission is the overwhelming majority.

In Europe, torque converters have never been much of a thing, historically because of the small engines (at the low end, and demand for control and responsiveness at the high end), and more recently for efficiency reasons.

AT has been shooting up in popularity (in parts because gearings have been getting shorter which makes manual transmissions really annoying), but mostly on the back of DCTs, even at the low end e.g. these days it's pretty common to find a 6-speed DCT on a B-segment car, and C-segments getting 7 or 8-speed DCTs (AMTs sadly still survive at the lower end of AT, but I feel CVTs are eating their bacon, especially with progressive hybridation).

Modern lock-up torque converter transmissions have pretty much eliminated any efficiency gains from other designs, I believe everyone is moving to them even in Europe. I know here in the US many care from various manufacturer are coming with German designed ZF transmissions including the 9 speed in my Honda Pilot.



> In Europe, torque converters have never been much of a thing, historically because of the small engines [...] and more recently for efficiency reasons.

The small-engine problem was basically just another aspect of the efficiency problem: Torque converters used to be so inefficient that if you had very little power to begin with, close to none at all got through to the drive wheels.

They also do it for simplicity (reliability and less maintenance). A that scale, a mechanical drivetrain to all those wheels would be much more complicated than electric motors and cables.

There are hydraulic trains, they have poor running efficiency worse then electric. My understanding is in order to get a large train moving with steel on steel requires very precise traction control that lends it self to electric, while doing a mechanical coupling for cruise speed would be very complicated and expensive so they eat the efficiency loss there. They are also saving brake wear using the traction brakes which just run heat strips on the top of the train.

> electric motors are better for that usage, since that is what they use

Except those that actually use hydraulic couplings or torque converters, which is the cheaper and lighter solution compared to electric, albeit less efficient.

BMW i3 has an optional range extender Which is a motorcycle engine that only generates power.

Nissan e-power cars are hybrids that you can't plug in, but all power is generated by the ICE and then fed to the electric motor before it gets to the wheels.

I guess I think the i3 doesn't count since its only a 34hp engine so it can't really power the car in normal driving (without sitting and letting the battery recharge for a while) its just a range extender but yes it is technically a series hybrid. It also has pretty poor efficiency using gas (31 mpg combined)

The Nissan e-power is interesting, having not dug deep into it before, it looked like it obviously had a engine to wheel connection through the transfer case. Thought they where pulling a Chevy and claiming series hybrid when actually parallel, but just digging deeper its very strange the engine output and drive gear are separated only by MM as though they where leaving the design open to a mechanical connection. Never the less its difficult to get highway MPG numbers on it which is why most hybrids have a mechanical connection. Electric is more efficient stop and go city driving while steady state highway mechanical transfer is more efficient. Would like to see power loss numbers and highway mpg, my guess is on the highway the E-Power would lose out to a Prius.

You are mistaken about the i3. Some friends had one and I joined them on several road trips. Driving it long distances is no different than any other car. You just stop and fill up the tank every few hours. The range extender pops on as needed to keep the battery from fully drawing down. It's important to remember it doesn't actually take that much horsepower to just keep a car putting along at reasonable speeds. Even economy engines are oversized compared to that due to the need for good low end torque in start/stop conditions. But a hybrid solves that the other way.

Although less of an issue than I thought there are driving conditions where it cannot maintain speed due to lack of power, it was a big enough issue for class action lawsuit to be filed over it dropping to 45mph on the highway:


And again my main point was on efficiency, I would hope you agree 31mpg combined is extremely poor for a 34hp engine when much larger engines get much better mileage and have no issues with providing power when needed without battery assistance.

It doesn't need much hp to maintain highway speed (probably around 34hp on flat ground) which means the engine is running full throttle to do so, vs running in it most efficient range which is not where it puts out peak power.

The e-power Nissan has a full size engine, but a small battery, although would be interesting to compare it drivetrains weight with a standard one, having both a electric motor and generator of equal size is typically much heavier than a transmission.

I have a BMW diesel sedan, and have averaged 56 mpg over the past 14 months. Starting from a dead stop is nearly all the fuel consumption, like an Atlas rocket. On flat ground, the car gets 70-90 mpg, and if not driving like an idiot up small hills, it's in the 50s.

My commute is very stop-go, otherwise I think the car would be in the upper-60s all the time. I've often wondered if there was a way to have an electric motor handle the movement of the car up to about 15 mph.

Which one do you have? That's remarkable fuel economy for city driving, and better than the older Volkswagens I've been considering.

It's a 328D. My earlier diesels were VW TDIs, a Jetta Wagon and then sedan. The BMW gets far better mileage, and the TDI sedan was one that was found to be in violation of emissions and was purchased back from me.

25hp is enough to propel a very non-aerodynamic scooter (it's about what most 250cc models produce), with the rider sitting up in the wind blast, at a continuous 130kph. An car can be more efficient at cutting through the air.

What it's not enough for is much acceleration at those speeds. But a battery buffer would help hugely, as long as you aren't driving very sportily for an extended period of time.

> There are no production series hybrids I am of aware of in automobiles

Would not the Chevy Volt count? The gasoline engine only serves as a generator, it is not mechanically connected to the wheels.

> The gasoline engine only serves as a generator, it is not mechanically connected to the wheels.

The Volt can operate in a series hybrid mode, but there is also a mechanical connection that is engaged at certain speeds where it's more efficient to just direct drive rather than double-convert the energy from the combustion engine.

> There are no production series hybrids I am of aware of in automobiles

The Chevy Volt was the first production series hybrid.

> Regarding efficiency

Air resistance is the main factor only if you’re not actively stopping/restarting/braking as needed. In urban or semi-urban courses adjusting speed and stopping at red lights is par for the course.

Also efficiency is really about effort. 95% is low for a bike, but even taking a 10% difference compared to the chainless one: for the effort you’d put to ride 10km with a chain, you’ll only be around 9km chainless.

As you say it might not matter if you’re only going for groceries at 1 or 2km of your home. That’s a different story if you use it to commute or plan on longer trips (now it could be seen as a handicap to get more exercise depending on the target customer)

People merrily put up with inefficient bikes for all sorts of reasons. If the only objective was to get between two points with minimal energy expenditure, everyone would be riding road bikes. Clearly other factors come into play.

I have a track bike that's light as a feather, a cyclocross bike rigged with single rear derailed that's barely heavier, and a dutch style city bike that weighs like 4x the track bike. Guess which one gets ridden the most?

I don't know. Which do you put your toddler on when you need to go shopping?

Why, the single gear of course :D

I think that was the point

The current trend is to get away from mechanical simplicity to have electric engines cover for these inefficiencies.

So, yes, people do care and want solutions to these trade-offs, and the current landscape already has decent answers to these concerns. IMO this chainless one would need to prove itself to be a lot better on some other dimension than pure maintenance.

Belt drives are very appealing on a pure maintenance basis. They seem worse on almost other dimension except possibly lower risk of a chainring tattoo (though perhaps I'm missing something).

Extra weight from the batteries and electronics could be a bigger efficiency issue.

Durability would be my bigger concern.

Perhaps this would be a good fit for something like an electrified velomobile.

I think it could in theory be more reliable and require less maintenance than a traditional drivetrain, but it's definitely going to be 7-10% less efficient than a good chain system.

> This motor is going to have to hold that force statically and have a control system with sufficient power and bandwidth to emulate the familiar feeling of the ground.

I understand that consumers will demand it, but ignoring that - what if they didn't simulate that resistance? Is there a functional need for it? What would it feel like? It's hard to imagine.

A chain is 95% efficient in theory, but in practice they are usually a lot less.

In theory a chain can be up to 99% efficient. In practice, a clean modern chain can easily hit 95%. Chains are by far the most efficient drivetrain.

How do belt drives stack up? I see more ebikes going belt drive which makes me think theres some efficiency gain there or at least its able to handle more torque perhaps.

Belt drives are less efficient than a well maintained chain but more efficient than a poorly maintained chain. E-bikes are less concerned about maximal efficiency and more concerned about ease of use. Belt drives don't need cleaning and greasing like chains do.

ebikes go belt drive for ease of maintenance and replacement, not for efficiency (they care much less since the motor compensates).

The elasticity of the belt could also be an advantage if the engine is at the pedals (rather than in the hub), as it would allow for a less smooth engine programming, using the belt’s elasticity to absorb some of the harshness.

> The Free Drive system works by converting the rider’s pedaling power into electrical energy via a small generator housed between the pedals. It then delivers this energy to the rear wheel (or wheels) via cables strung inside or outside the frame of the bike, rather than sending it mechanically through a chain or belt. Excess energy created by pedaling is fed back into the battery. The end result is a power system with fewer moving parts to complicate construction.


So... a generator, an electric motor, few microcontrollers, bunch of sensors, a battery with it's own management circuits, wires, a bunch of code running on everything... is supposed to be "less complicated" than a chain and two (sometimes more) sprockets? Or did I misunderstand the press release?

I don’t know about you, but I pretty regularly have my chain derail on my bike (at least a few times per season). Then I end up with grease on my fingers as I fix it which can be a big inconvenience.

Do you regularly run out of power on your chain powered bike though? I get the feeling this system requires a battery (it even mentions battery in the article), and most of the actual power comes from that. The question is, will the generator make enough power to get you moving if the battery is totally flat? I kinda feel like no, you'd have to sit there and pedal it just to charge it up, and then it would move you a few feet. Or you're going to be pedaling really fast, and moving nowhere.

They claim to be 5% less efficient than a chain. So with a dead battery, you'd be limited to a speed of 95% of what you could do on a normal bicycle. That's if you're already moving. If you were at a dead stop I imagine you'd need to pedal a few strokes to get enough charge to get enough torque to overcome the stiction.

Where do they claim that efficiency? I'm pretty skeptical... Chain+Derailleurs can be 98% efficient IIRC

Electric motors/generators are in the same range. Also I doubt they need to target the best possible chain bikes in order to be a viable solution.

Since it has a battery and the battery is (most likely) chargeable, a little loss of efficiency is inconsequential. On the other hand, having an actual physical chain on an electric bike seems indeed unnecessary - at least for less sporty applications, where things like "feel" or precise power delivery are not important. The end result will be a simpler, more reliable bike. Not to mention better looking.

I googled Schaeffer Free Drive efficiency.

A Tesla motor is 98.5% efficient, so a generator and motor combo being 93% efficient does not sound like a stretch.

You need to get your derailleurs adjusted. I bike almost every day and it never happens to me, even when I do inadvisable hard shifts.

Been there. If you're riding a 1x drivetrain, a narrow-wide chainring will greatly reduce chain drops. A conventional chainring is designed for the chain to be thrown off easily, and the front derailer keeps it there. On a 1x system, the ring can be designed differently.

Hey thanks for this. I realized one of my bikes has been built wrong since I got it 15 years ago due to this comment.

Not really a normal thing;tTake your bicycle to a bicycle shop for diagnosis and repair. It make take a follow-up visit. Sometimes issues around dropping chains can be a bit complex to diagnose.

It only happens once a season or less for me and is almost always due to adjustment problems. And 90% of the time I can get it back on without touching the chain by using the front or rear derailleur.

If you're riding a single speed bike and the chain regularly derails, then that definitely sounds like a chain tension or alignment problem.

If your chain isn't too badly fallen off, you can usually just shift in the opposite direction of where it fell off and the derailleur will usually get the chain back on without having to get your fingers greasy.

Be careful when you do this so as to not apply too much power when you're trying to do this. If you're in a really high gear and it happens, you can get off, hold your rear tire off the ground and turn the pedals. Works 90% of the time unless you got your rear chain jammed between the cassette and the spoke.

Edit: You can very often see this if you watch professional bike racing. When the mechanics change a rear tire it will almost always knock the chain off of the front chainring. They'll just shift into the proper direct (low if it's off the high side and high if it's off the low side) and then spin the cranks and the chain will come back on.

I've gone entire seasons of hard mountain biking and touring without having my chain come off. I think your bike might need an adjustment or maybe you have a worn chainring.

1. Check your limit screws. Park Tool has nice Youtube videos for adjusting these.

2. If it's the rear derailleur, it may be bent (especially if your limit screws from step 1 are maxed out). You need a derailleur alignment gauge (or just take it to a shop) to verify.

If it bothers you that much the solution is a belt drive, which is still more efficient than a generator.

Wait until your chain gets ransomware

It's really catastrophic when the drive chain's watchdog timer has to reset it because some function went awry.

Not a normal thing. My chain never goes out. Only on janky beater bikes it did. To the bike shop you go.

Tuck a pair of nitrile gloves under the saddle or into your toolkit. No greasy fingers.

Have you seen a high-end bike? The gears have batteries, electric motors, microcontrollers, and sensors. You'll also find wireless power meters, GPS, and even radar sensors for traffic on a contemporary bike.

You get the idea. At the leading edge, bicycles are already extremely complicated. If we're comparing this idea to a bicycle from 2002, it's complicated. But I'm not sure that it's that complicated compared to the kind of bike you'd find in the roof rack of an Audi S4 Wagon :-)

I've seen them, even rode some bikes with electric shifters, but still don't want the complexity. One of my bikes is over 20 years old with the same Ultegra shifters, I haven't done any maintenance other than lubing once a season and they still work fine. I'd be surprised if the battery on electric shifters lasts that long.

You're describing a high-end e-bike, which is basically an e-motorcycle with less regulation.

A high-end regular-bike has electronic shifting and a power meter and that's about it. And you're talking many thousands of dollars worth of road bike. It's like seeing a Ferrari or a Lambo rolling down city streets. They exist but they're rare.

Yes, but we're discussing technology that will also be rare. It's not like these chainless electric drive systems will appear on big-box sporting goods store bicycles, so why compare it to a mass-market affordable bike's technology?

I think it's reasonable to compare a leading edge technology to a leading-edge existing product. Otherwise, it's like it's 2007 and we're complaining that this newfangled iPhone thngie can run out of battery in a day, while my POTS telephone works even in a blackout.

Yes, true, but apples, oranges.

Not necessarily. Shimano Di2 electronic shifting was introduced all the way back in 2001 and has been among the standards for high-end road bikes for many years now. It has all of those components. The electric motor doesn't have to drive the wheel.

One comparison is to e-bikes.

Which often have most of that and a chain too.

But also, e-bikes can let you ditch gears, which add complication.

But I think the main selling point is that you can seperate the sitting/cycling position from the driven wheel in interesting ways.

The selling point might be to finally have an "automatic" bike. Standard derailleurs are basically like having a stick shift on a car. The rider has to be trained on how to use them and they will make errors like leaving the gear setting too high when going on an uphill and then being unable to downshift because they aren't turning the pedals fast enough due to the high gear ratio.

This is literally just get on and start pedaling and the bike figures out the rest. It even makes e-bike controls super easy, since you can just set a speed and start pedaling and it will supply just enough juice to make up the difference.

The obvious downside is that it's going to be less efficient than a chain drive, because nothing beats a chain drive. But if you basically get the benefits of an e-bike for free then the efficiency loss isn't a big deal for the rider.

There's at least one existing hub drive ebike that uses this model.

It has a single gear, you can spin the pedals backwards to engage full regen, but the secret is that its usually doing a little regen which it stores and uses to help on hills and when accelerating from a standstill.

I am surprised automatic shifting technology doesn't exist. Wireless electronic shifters are readily available now, in addition to power meters. We have devices to shift and devices to tell us when to shift. All that is needed is to marry the two together.

It does exist, in the form of a CVT and controller invented by a company called NuVinci. I built a cruiser bike with the manual CVT version years ago. It worked but the hub was very heavy. If you were stopped in high gear and you pushed down on the pedal really hard to start going it would occasionally slip. Once you got going it was pleasant to use but don't expect to win any races.

They must not have sold very many of them because Fallbrook-NuVinci went into chapeter 11 and enviolo bought the tech.


The Vanmoof bikes have automatic shifters. I don’t like them because they are unpredictable, and you feel the difference. It’s jarring. The Cowboy is single speed, which works pretty well because of the motor. That’s my preferred configuration. But at high speeds it does feel like I’m at a spinning class, and it sounds like this system could fix that.

Lets be real, the selling point of this nonsense is that you can ride your electric motorcycle on a bike path and pretend it is a bike because it has pedals.

Personally, I don't get the hate over people using e-bikes on bike trails. While I am still fully pedal powered I have no problem with other people using e-bikes. Especially if the alternative is for that person to drive somewhere in a car.

Maybe in other areas people are doing 60kph on bike trails or something, but that's not what I see around me. They seem to top out around 20kph or so, which makes them basically just bikes as far as traffic flow is concerned. Sure they will zoom past you on the uphills, but who cares? It's not hurting me or anybody else.

There are lots of overpowered e-bikes with chain + gears, where the power going through the pedals is maybe 5 % of total power, and the rest is electric. This doesn't change or bring anything new to the table.

Far more complicated that a traditional bike. But compared to e-bikes, this avoids a lot of the complex mechanical bits.

I've had this system in mind for years, I think for cargo bikes with long chains, this will be a clear winner.

Bike drivetrains can require a fair amount of maintenance. Also a normal chain derailler system requires at least four sprockets, two in the derailler, one in the wheel, and one in the bottom bracket. Anything less is likely to be less reliable.

You can expect to change the chain every 5000km. In theory a fully electric drivetrain could last forever.

All I do is lube mine and its been a gem for years now. People preach that there is this whole maintenance regimen and you have to be this hobbiest watchmaker to deal with bike gear ratios and such and such, yet most bikes just sit in the garage with the lube that came on them from the factory and work fine. Go to a college and see all the 50 year old 10 speeds lined up in the racks. If you have a 50 year old piece of hardware in such high use that doesn't seem like its this unreliable untrusty system to me, quite the opposite. Especially compared to like anything else in transportation, like a car that might incur thousands in repairs over its life. How many 50 year old cars are parked at the college? How many cars are even over 25? Even a skateboard is less reliable; take a skateboard out in the rain and the bearings are fouled and the deck may even warp. You can even get bike tires that make flat tires practically obsolete short of hammering in a nail because they use kevlar like a bulletproof vest. Literally bulletproof tech.

Your phone is way more complicated than your bike and the hardware probably fails less often.

I've had my bike 10 years. How can the hardware fail on a bike? Throwing it off a cliff? Meanwhile, the laundry list of hardware issues I could list on my phone probably affects every component in the stack over my decade of owning various smartphones. Each dying due to some faulty hardware issue, such as the radio giving out and dropping calls or the jacks getting loose or various other issues.

My phone doesn't sit out in the sun/rain/snow at a bike rack and doesn't have any substantial moving parts.

I think I didn't articulate my point well. It's that complexity isn't a 1 to 1 mapping to reliability. There is plenty of complex equipment that fits your criteria. My pool pump is full of electronics, has moving parts, sits in inclement weather, and lasts for years. There is equipment that can sit near the combustion chamber of a rocket experiencing extreme g forces, pressure, and heat and survive. It can be engineered.

Phones tend to get pretty nice treatment compared to bikes. Bikes are constantly stepped on and rubbed against pavement.

Because if something goes wrong with your phone you probably won't be able to fix it. Compared to a bike where most anything can be repaired.

It's a single unit, you just replace it, like you would a broken gear. You can't fix a broken gear unless you weld it.

Idk, if you tried to use your phone as a bike I bet it would fail pretty quickly.

I don't get your argument. What I'm saying is that complexity isn't necessarily directly correlated to reliability.

Sure, but a hammer is both simpler and much more reliable than a phone. My point is that it is meaningless to compare items with completely different functions (I would say apples and oranges, but it is more like apples and rocks).

pedal by wire

No different from how most diesel trains work.

Totally different application/need though. Locomotives need to generate enormous amounts of force, for long periods of time, regardless of speed. Often at zero speed.

That capability is much more important than the efficiency loss of the generator/motor powertrain.

Such capability is not necessary on a bicycle, where efficiency is extremely important.

Bikes handle much differently from diesel trains. This may work for some but I can see it occupying some kind of uncanny valley in the riders' psyche for a while.

You ever ride in one of those bar on wheels things you find in touristy places?

It's got "barstools" along both sides and the passengers/revelers are supposed to turn the pedals under their seats while the bartender/driver steers.

In reality, the pedaling charges the battery somewhat but the vast majority of the battery power comes from being plugged in at a charger before the ride. You can't actually rely on a bunch of drunk sods to keep it moving.

> Bikes handle much differently from diesel trains.

So, like, no nudging the right handlebar forward to turn right? Bummer!

I would like to have a bike with this drive system for doing structured training workouts outdoors. Cyclists and triathletes often do workouts involving multiple steps with a prescribed power output, for example 10 one minute intervals at 300 W with two minutes of recovery at 140 W plus 15 minutes of warm up and cool down at 170 W. It's simple to execute that workout on an indoor smart trainer in erg mode since it will dynamically adjust the resistance to hit the exact target power, but even with a game like Zwift indoor training is super boring. You can do the same workout outdoors on a real bike, but then you have to keep watching the power meter to stay in the target range which is distracting and somewhat of a safety hazard. So this system could offer the best of both worlds if the target power is externally controllable through ANT+. Of course that would probably be only a small niche market.

Hi, I remember that an app similar to what you are talking about were developed for the specialized evo ebikes. You could set a fixed power output and if you went higher the motor would come in to help you stay in the power You choose. Pretty cool.

I see your point, and I can see some application in group rides, social events with a less fit SO, but 1) in TT, or even in any regular bike climbing, holding constant power is an important skill to master. Try doing a 45min varying gradient climb at 5-15% below FTP. It is not only your legs, but also your brain anticipating the gear shifts. Give the Alpe du Zwift a try, turn ERG mode off, and you'll understand what I mean. Variance in the power output wears you down. (that's why we use Normalised Power numbers for fatigue). You've to train that, not by doing3min on-1min off intervals, but by riding on real tarmac. Also, after few thousands hours of training you don't need to look at your head unit to know if your power is constant or not. 2) Power is not the only bottleneck for riding 300W, 400W, 450W for 10min, that where the handling skills of your bike come in play. That one of the reasons why we do 'speed training', cycling behind a buddy on a scooter/car. To get used to the cadence, braking, steering, ... Imagine participating in a race, riding 50+km/h without any fast cornering/breaking experience, that's a hazard! 3) Imagine doing an all-out interval session on TT bike, in a full aero position, and getting passed by an elderly on e-bike. Everyone will think "that guy is a bit weird", won't they? 4) Max. power output of this motor is 250W, I believe this to be an EU regulation. That is 20% less power on the rear wheel than the FTP of an average amateur racer. There are also speed regulations in EU, 25-28 km/h. That's certainly the Strava numbers to show off! You can make it a motorcycle/scooter and remove these limits, but then you'll need to wear motorcycle helmet in many EU countries. Also, you are no longer allowed on the bicycle lanes. 5) People who are serious enough about their cycling are generally quite serious on the bike position. For example, the Q-factor is the distance between your shoe cleats. Favero released power meter pedals for Shimano SDP-SL cleats, but added 1cm to the q-factor, just look at the flame war it caused. 6) Most people train on their 'race-day' bike. Some have a spare 'bad weather' bike. Now, you want to train on an additional training bike, with a different geometry to the race-day bike? Or are you expecting to have a TT bike with this drive train + a race day TT-bike? TT-bike will run into 5-15k USD, I might like my hobby, but won't put extra 5-15k for this nice to have feature.

I think 'niche market' is an underestimation, since there are already assisted racing/MTB bikes (limited to 28km/h), just slap power meter pedals on and you're ready to go.

Northern climate bike commuter here. I am getting tired of commuting in the snow and rain. I want a 'winter bike' that is fully enclosed, allows me to use my bike lane to work, and get my exercise without being covered in sleet, rain, or snow when I get home.

When I was younger, the machismo of winter/rain commuting was a fun brag at parties. Now, when I look out the window after a long day at work. I want my exercise miles minus the macho.

This could help me get that winter bike.

You can look up velomobiles. They are quite expensive though.

Sounds like podbike https://www.podbike.com/

The enclosure is a lot of mass. Easier to get the same sheltering effect with better rain/snow clothing that you can just strip off and shove in a pannier or something.

> The enclosure is a lot of mass.

It's also a lot of sail surface.

Isn't it more aerodynamic than a human on a bicycle?

A rider on a regular bike has poor aerodynamics in every direction as their cross section is not that much different with an open frame and spoke wheels. But a sleek enclosure would be like a wing and have very different forces depending on the angle of attack against the wind. A gusty cross wind could much more easily push the vehicle across lanes or even knock them over...

No, not really. Not at the speeds most cyclists, especially commuters, travel at. You gain quite a lot more by going recumbent when it comes to drag. In fact, a recumbent bike, for the same effort, is a lot faster. Putting a fairing on one of those is another story.

> Networked, flexible, sustainable, and environmentally friendly

Adding batteries, more waste and breaking points is now "environmentally friendly" because "electricity"

Electric is generally a lot better than driving a scooter with an ice around.

But a "dumb" bike is miles better than an electric one in term of ecology.

Replacing a ICE scooter is an electric bike is positive, replacing a regular bike with an electric one is a net negative. The argument only works if it replaces a more polluting option. Just like a Tesla can be a net negative depending on what it replaces

It depends on how you do your accounting. The electric bike is more efficient than the pushbike, just as the pushbike + human is more efficient than the pushbike alone.

The carbon cost of a joule of food is higher than a joule of electricity, and joules are what get you where you're going.

You can make a case that it's nice exercise you needed anyway; I like bikes as well. But the claim you're making isn't obviously correct and I would say it's more false than true.

It also doesn't account for reduced car use due to having a more convenient mode of transportation than a standard pushbike.

Thats fair, I tend to approach this problem as the way out is to innovate more. Unless we hit total disaster and civilization collapses there is no way to coordinate humanity to think about net impact at scale. So when I see an electric bike I see it as one possible less car on the road which to me is in the right direction.

The net effect of ebikes isn't going to be for existing cyclists to buy batteries.

Like what is your point? That we aren't on a path to a perfect society? Do you think complaining is going to change that?

I'd be more interested as an alternative to chain or belt drives but I'm not sure how you incorporate gearing without power electronics. Chains can be up to 98% efficient and I'd prefer not to throw away 10% just to eliminate gears.

It's worth noting the "up to" part. Well maintained and cleaned chains are effective, but I would guess that a lot of commuter bikes have dirty, rusty, and worn chains that lose with a lot of efficiency. Couple that with poorly maintained and adjusted gears and alternatives might not be far off in efficiency if they're a closed system with less maintenance requirements.

You also have to take into account the extra weight of having a dynamo + a motor, I doubt it's more advantageous than a chain on a classic bicycle.

For an electric bicycle though, I see it as a really good contender

Yes, this particular system is probably far from efficient enough for regular bikes, this is more for larger transportation bikes and similar (like this example [0]). Having a chain on those limits their construction options, and they're already electric anyway.

I was referring to different types of gears and power transfer like driveshafts or hydraulics [1], if you can make it perform at a consistent 90% efficiency it might beat chain drive in practice even though it shouldn't in theory.

[0] https://youtu.be/N02KMeOkevI?t=81

[1] https://www.youtube.com/watch?v=NTB7XOQA-XQ

I'm not sure how this system works, but it seems like adding in battery power might be useful.

I'm assuming it was suggested to replace a chain by what is essentially a dynamo + motor.

You pedal to generate energy, and move that energy to the wheel through wire and another motor. (I'm not saying this is a good idea, just describing what I understood).

I'm curious what sort of efficiency you can expect on a ridiculous system like that.

I don't have a link handy but the difference between a perfectly maintained chain and a dry, poor condition chain is not as significant as people think.

Also, many, many transportation bicycles have chain cases, which is much simpler than a motor-generator system.

Let's not forget about belt drives as well which are becoming more popular due to no need for maintenance while still having great efficiency.

This system enables regenerative braking on bicycles. That means you also gain a lot...

What I would love to see though would be this, applied to a recumbent or a velomobile. Those often have atypically long chains.

There already were regenerative electric bikes. The first-generation Specialized Turbo had regeneration linked to the brake levers, and nobody cared. Not worth the complexity.

Later Specialized bikes have a regen mode you can engage manually but it's still basically never worth it.

Why isn't it worth it? It seems like free energy. Or is the weight a problem?

Most of the energy is being lost to wind resistance and rolling resistance and parasitic friction. The fraction you can regain by recovering potential energy down hills, multiplied by the round-trip generator-motor efficiency, just isn’t relevant to the overall experience.

Thanks. A couple thoughts: I read somewhere that rolling resistance didn't add up to much, but maybe I need to recheck that. Also, I wonder if it's different for city riding, where riders need to stop every block or two and start again.

I doubt it. Due to weight balance, most of the braking force is on the front wheel. This system only drives the rear wheel. Any regeneration gains would be minimal.

Not at all true, I built a dual suspension bike with a direct drive hub in the rear. I have it setup to regen, and I can use that to stop without touching my mechanical brakes for 95% of my riding. You generally only recover 5-10% over the course of a ride, but I am in FL with no hills, your results may vary!

That only comes into play under hard braking. In most cases (gentle deceleration) there isn't much weight transfer: the rear wheel is fine for light deceleration.

The controller on my e-bike indicates ~120 watts of regen from the rear-wheel. I'm pretty sure that's limited to increase life of the battery; there's plenty more available under heavy breaking.

It would be awesome for replacing the drag brake on a tandem for controlling speed on long descents.

This is a niche in a niche though.

I had a drum brake on my tandem for this very purpose. It wasn't even hooked up to a brake lever, it was wired to an old three-speed shift lever so you could set the amount of drag between none, some, and more.

My tandem doesn’t have the brake, but it does have the braze ons for it. I haven’t bothered because I’m not doing loaded touring or mountain routes.

Would be a game-changer for urban riding, being able to quickly reclaim energy after a stop.

Regenerative braking makes a lot of sense on cars because cars are not very weight-sensitive and more weight works in your favor for regen braking.

Regen braking makes much less sense on bikes because bikes are weight-sensitive.


That blog post is pretty misguided, IMO. I'll assume the math is right, and the potential gain from regen braking is small. But it's free!

> The way regenerative braking is implemented is to have the motor continuously engaged.

This is wrong. Regen braking only requires the motor be engaged _when braking_. There's no reason you can't coast when you want to coast.

It's certainly opinionated, I don't view 50-100 stops in ~30 miles as at all unreasonable.

Something which is rarely done but easily could be for a bike is regenerating first into a couple supercapacitors, which are 99% efficient and fill and discharge as fast as you can push current. They don't have the capacity to be as useful in cars, for a bike they can also provide a nice kick to overcome starting torque. The downside is it's another two drink can's worth of volume to add somewhere on the bike.

You are absolutely right about it being small, but free.

Do keep in mind, a direct drive is always "engaged" to some extent by the cogging force. An advanced controller can make a direct drives "freewheel", but it actually takes power to do that. Totally worth budgeting for a slightly larger battery to allow a simpler overall design.

Now if you say, had a mid-drive bike, or an internally geared hub, there are some complicated mechanical systems you can put in place to lock the freewheeling mechanisms in the gearing when you want to apply the brakes, but I haven't seen anyone design that yet.

Does cogging actually absorb energy though? (When averaged out)

Regeneration on a bicycle is only possible on hub motors, which usually have a planetary gearset to have acceptable low-end torque, and that gearset has significant drag.

Mid-drive bicycles are able to leverage the freewheel, and thus have very high coasting efficiency.

You don't really seem to know much about this. It's odd that you seem to have formed strong opinions on said subject.

There is fierce competition in the industry for efficiency/range/price...not really sure why you think you can just casually stroll in and go "well DUH, folks, just do..."

With these, you could store energy by pedaling, even when going downhill. If you pedal downhill with a chain, that's just wasted energy.

Unless you downshift and peddle hard to gain a lot of speed :-)

You're joking, but to answer the question seriously, the problem with slamming it into your biggest gear and pedalling downhill is that wind resistance increases with the square of your velocity.

When going downhill, you are going faster thanks to gravity. A small increase in speed requires a disproportionally large increase in effort. Hand-waving over the inefficiencies involved in an electric power train, you are far better off pedalling and storing the energy on the downhill, then "spending" that stored energy on the flats or even saving it for the next uphill.

With respect to going downhill faster with this system, I doubt it can do anything useful in a straight line descent, but in a long, switchback descent of the type seen in the big Grand Tour races on mountain stages, cyclists need to brake into the switchbacks and accelerate out of them.

Regenerative braking followed by assisted pedalling out of the corners would be a huge win.

> Regenerative braking followed by assisted pedalling out of the corners would be a huge win.

...Until the next ascent, where you have to haul the weight of the battery and what amounts to two electric motors uphill...

Although it seems sacrilegious to imagine energy-recovery units in UCI bicycle racing, innovation is strongly influenced by the big manufacturers who are trying to sell bikes.

It's unlikely there will be a UCI-sanctioned ERU any time soon, but if there was, the key to adoption would probably be the UCI minimum weight regulations. We are now at the point where high-end bikes often need weights added to meet the minimum, and if the rest of the bike ever gets light enough, it could be possible to add an ERU without compromising the total weight for climbing.

But that would require breakthroughs in the culture of bicycle racing and multiple technologies.

>"...The problem with slamming it into your biggest gear and pedalling downhill is that wind resistance increases with the square of your velocity."

-- Are you much of a biker? (I put on around ~1,000 miles a month on a 29" mtn bike.

Knobby tires and all.

At no point ever have I worried about wind resistance - and im not a "shave my legs tour de france" (my brother is, but hes an ultra athelete-type-A Doctor) type that worries about my grams per component, corporate spandex or $12,000 week-end ride.

So, while you may be "technically correct" you're commenting as "functionally illiterate"

I really don't want this to devolve into chest-pounding, but I have worked in the bike industry, and have raced on-and-off as an amateur in road, cyclocross, MTB, and duathlon dating back to the late 80s. I remember when the Ritchey P23 was the state-of-the-art in lightweight mountain biking, something I chuckled at when riding a 17 pound carbon Ibis a couple of decades later.

Wind resistance absolutely matters in every discipline of cycling, and you don't need to have been fooling around with bikes for forty years to know that. If it didn't matter, what are all those triathletes and time trialists doing with aero bars, flat backs, and disc wheels?

It matters in MTB as well. The optimal position for efficiently generating power on a bicycle is actually quite upright, you can see this if you look at pictures of people riding "roller races," they usually flipped their handlebars up so they could be much more upright than when riding on a track.

XC MTBs have much lower bars than would be most efficient on rollers without wind resistance, and that's because the lower position generates less drag, knobby tires and all.

Even if you aren't riding a time trial or racing XC on an MTB, knowing where to expend your energy and where to save it matters greatly. If your daily commute involves hills, you will work less and arrive sooner if you don't try to crank your max while descending, and save your efforts for climbing.

That's just math and physics.


Also, please, I'm not upset at any random internet person using insulting language, but it is not constructive for our community to go around suggesting other users are "functionally illiterate."

pwnd. me. your comment.


and thank you for the response, proving your functional literacy :-)


Anyway, when you close with just "Math/Physics"

Its a great comment, I just dont, on my personal, take ANY of that into account, as I am there for the ride and not to best anything.

But I LOVE everything about what you said and do.

I need to know you

Here’s the thing: There’s talking about cycling, and there’s cycling. They each have a role: Talking about cycling is how we share knowledge and build a culture and an industry to serve it, and grow knowledge over the generations.

But it’s still not cycling. For that you have to throw a leg over the top tube and go out and suffer, and fall, and get up, and bend a bent hanger back to fix it after the fall.

And ride, and ride, and ride. And wear a shit-eating grin so hard that your face hurts, but you can’t stop grinning in the wind, because riding feels so damn good.

If you ride, that’s all I need to know. See you on a trail someday, maybe.

Mountain bike descents and road bike descents are dramatically different things.

Find me a road bike with an electric motor.

They exist, and that's why UCI scan bikes at the race start line for electric motors, it's a serious enough problem that it has a name, "Motor doping." Famously, Femke van den Driessche received a six-year ban after being caught motor-doping in cyclocross.


If you don't want to cheat with a motor, bicycle stores sell all kinds of road bikes with pedal assist for non-competitive riding. Trek, for example, make several e-assist versions of their Domane road bikes:


Where newsletter and how subscibe

I live on a bike. It's my primary mode of transportation. This could be a very interesting choice. Getting around is not about speed. It's about ease. If this system can get people around more easily then it's worth pursuing.

There are bicycles with chains, belt drives, pedal assist electric, and throttle electric. I would be interested to know if this system could work without plugging in, get me up a steep climb, and do it cheaply. I use chain drives but belt seems to be the winner at the moment.

Take away the chain and you are on the slippery slope to electric motorcycle. This thing is clearly meant for an electric moped. Up the batter/motor power a little and the human input becomes irrelevant. You are then driving an electric motorcycle with some pedals for topping up the battery. Absent the chain-to-pedal drive, the only difference become e-bike and e-motorcycle is some arbitrary line about battery/motor capacity.

Some of the technology choices are driven by: 1) Regulatory distinctions between bikes and motorcycles (and scooters, etc.), how they are licensed, where they can be used. 2) Providing the experience of low intensity exercise.

Don't remember where I read it, but e-bikes are becoming prevalent enough in some parts of Europe, that there is a bit of a regulatory backlash going on, for instance requiring a bike to get at least 1/2 of its power from the rider.

e-bikes around here (Boston) seem to be mostly motorcycles with vestigal pedals. If the rider stopped pedaling, I'm not sure there would be any visible change in vehicle behavior. I may not like it but I prefer it to having more cars on the road.

It's interesting, because this exact transition (bike+motor assist -> motorbike) has been seen before. In Britain we had the category of 'moped', complete with vestigial pedals, before they were subsumed into the small motorbikes (limit: 50cc engine).

The difference is that now the motor is electric, not a small petrol engine. Will be interesting to see if we end up treating them (again) as small motorbikes, or as bikes with assist (so no compulsory helmet, insurance, or registration).

Edit: around here (Liverpool) I see them styled both ways. Some look like motor scooters with pedals, others pedal bikes with a motor. Seems to be fashion rather than functionality that determines which.

At least here in the states, the "moped" was not meant to be pedaled while in motion. It was for starting the motor. The gear ratio was too low to be useful while riding.

The Honda Spree, with electric starter and centrifugal clutch, put an end to mopeds.

In my locale, they are more like small motorcycles without brakes.

Actually, most of the e-cyclists are quite well behaved, but there does seem to be a tendency among the beginners to maneuver through things at speed, that a conventional cyclist would slow down for. My guess is they haven't developed a sense for how far ahead of themselves they actually need to be paying attention.

But also, I admit that there's a certain bias here. There are fast and slow cyclists, and those who are polite or jerks. You don't notice the polite cyclists at all. You don't notice the slow jerks, because they're behind you. You only notice the fast jerks, on either electric or conventional bikes. So there's a bias towards thinking that fast riders are jerks.

>> without brakes

That is key. Brakes are largely unregulated on bicycles, at beyond laws saying that they need to have them. Standards for braking power and/or stopping distance would be fought tooth and nail. All those ultra-efficient bikes on slim tires wouldn't be possible if someone set minimum stopping distances.

Actually I was being sarcastic, always a bad idea on web forums, sorry. "Without brakes" meant that people don't slow down when they should. And to be charitable, I think it has to do with people learning what kind of situational awareness and reaction time are needed.

The US CPSC does regulate bike brakes. The latest generation of e-bikes actually have quite effective brakes, nearly entirely hydraulic discs.


Those are regs for sold bikes, not bikes on the street. It doesn't cover customized/user-modified bikes which are the vast majority of commuters these days.

There was a man in the UK recently charged with the death of a pedestrian. He was on a fixed-gear bike without front brakes. Causing death during illegal activity means homicide charges in many jurisdictions.

Aha. My state prohibits operating a bike without a brake, though the regulation is light on specifics. I read about that incident too.

Eh, the slow riders are jerks too :). At a red light, they'll cut the line and blow through the intersection. Then the light turns green and you overtake them until the next red light. Rinse and repeat until you get to work/home.

Not to excuse such behaviour, but it can be explained in part by minimising time sucking emissions which are typically most noxious when a ICE is accelerating. So these jerks wishing to get upwind of accelerating tail-pipes is rational, even if selfish.

There are plenty of people with throttle-less electric hybrids/mountain bikes as well, but they blend in with the other cyclists (until the road tips up hill).

Aren't ebikes just electric mopeds with less comfortable seats? I always heard moped was just a term coined by a journalist trying to describe a bike that had both a motor and pedals.

I like innovation like this, but I will never buy one. A chain or belt will always work when the battery is dead. Or something went wrong. I need a fail-safe mechanism, and this idea does not offer one.

Conventional mechanical drive trains are very efficient, but those with multiple gears can be fragile with exposed moving parts subject to damage and misalignment. Most of us who ride a lot have experienced broken shifter cables, bent derailleur hangers, etc. Hardly fail safe. A well engineered electric drive system could potentially be more robust and reliable, but obviously more expensive and less efficient.

I agree, but even the most sophisticated or rusty/bent/damage drivetrain is only unridable if a chain-link is broken and you don't have a chain-link tool to shorten it. Even with broken cables, you can select a gear by moving the chain and go forward.

  An escalator can never break: it can only become stairs. You
  should never see an Escalator Temporarily Out Of Order sign,
  just "Escalator Temporarily Stairs. Sorry for your convenience."
--Mitch Hedberg

This is a funny quote, but for those who don’t know it’s also wrong. Escalators can indeed break, and violently: https://gizmodo.com/catastrophic-escalator-failure-in-rome-s...

There's also the fact that an escalator makes for some terrible stairs — the steps are much taller then what would be comfortable (and allowed, AFAIK) for stairs.

I cannot count the number of times I trashed my gears. And the fail-safe is walking. (I assume you're using a fixie?)

Something is very wrong with your bike if you are frequently "trashing" your gears. I've been riding bikes for 30 years and have never had to walk due to drivetrain failure.

As long as the drivetrain is set up well initially, a traditional bike will go many thousands of miles on pretty minimal maintenance.

I suspect I'm just not taking good enough care - both while riding and with maintenance. Rusty bowden cables causing misalignment and the chain jumping into weird places… (I've also used some sketchy bikes. One literally had the rear derailleur bolted on because the mounting mechanism didn't fit the holes on the frame. I ended up completely twisting that one.)

Single speed transportation bicycles are extremely common where there aren't significant hills. Internal gear hubs have regained significant popularity, too - ranging from 2 to 11 speeds.

Not sure why you're "trashing your gears" so regularly. Older cassettes/chains didn't like being shifted under significant power but anything made in the last 20+ years by SRAM or Shimano or Campagnolo really shouldn't have a problem being shifted while full-on sprinting.

I think we may have different definitions of what significant power means. I managed to break the metal rod inside my plastic pedals one time… But yeah, I used an 8 speed hub gear with a toothed belt for ~7 years, which I don't see breaking any time soon. (Though I did have to walk home a few times due to other reasons, e.g. clogging the gears up with snow or flat tires.) But it doesn't switch gears when under power, and if you start giving power while it hasn't fully switched yet, it makes very unhealthy sounds.

This is a generator, a dead battery is not a problem. Nothing in this setup is less likely to break than your chain or a derailleur -- I've had those happen to me a couple of times each.

I understand what you are saying, but a generator capable of providing enough power to propel the bike without a battery doesn't exist yet. If that was the case, the battery would not be needed.

It is however a lot easier to cart around a chain tool and a replacement chain than a replacement generator.

Which I do on longer rides, but they only seem to break on casual rides when I don't have them.

But I've also broken frames and taco'd wheels. Should I bring spares for those?

That generator can probably do a million miles.

You must be putting your bikes through hell and never maintaining them. There are no instances where a chain should break if properly maintained and replaced at the end of its life.

It's happened twice. Once was on a really old chain. The other I'm not as sure. I'm a heavy guy. If I use my weight I can put a lot of torque on the pedal.

Classic fact from my old college E&M (electricity and magnetism) course:

"A series wound electric motor has infinite torque at stall"

Okay, the electric connection between the pedals and the drive wheel should act as essentially as a perfect transmission, that is, with infinitely many gears (except for the 5% or whatever is lost in efficiency). And if the wheel motor is "series wound", then should have the "infinite torque" when starting from a dead stop. That infinite torque could be nice to have when going up a steep hill -- e.g., for the last 10 speed bike I had, the lowest gear was still not low enough to let me pedal up my steep driveway and, instead, I had to walk my bicycle up that hill.

Also, for that bike, the highest gear was not high enough -- on the course I was using, there was a long hill, and in the highest gear before I got to the top I was pedaling as fast as I could and wanted a still higher gear.

Sooooo, for something better, if the constant torque generator had a resistance adjustment, then just increase the resistance a little, let me pedal at the same RPM as before, and get up the hill faster; that is, I would be pedaling with my maximum power and the infinite gearing would move the bike at the maximum speed for that power, e.g., the power needed for the friction and air resistance for that maximum speed.

Constant torque at the generator side and infinitely many gears connecting to the drive wheel -- NICE!

I cannot see why anyone would prefer this for a bicycle that is at all shaped like a normal bike. An integrated mid-drive ebike is just all around much more efficient.

Direct hub motors for large (bike size) diameter wheels frankly suck at this point in time. You need to have much more copper to efficiently drive the motor at the power levels required, because RPM is so much lower with hub motors. Not being able to use RPM to your advantage is such a huge efficiency hog at low speed acceleration, which is a lot of how people use ebikes.

If this is using an internally geared hub, my point is moot; however that comes with it's own drawbacks. It's incredibly hard to get heat out of a geared hub motor efficiently. Your motor is encased inside the hub, with no direct connection for heat to escape, and you generally have way less copper available to heat soak.

Mid drives get the advantage of mechanical gearing, and can be built in such a way to allow very little heat generation to begin with, but you also can easily cool something in the frame compared to a spinning hub.

> I cannot see why anyone would prefer this for a bicycle that is at all shaped like a normal bike.

There are already e-bikes out there with a design that puts a motor into the bottom bracket; so there are already frames for this.

That's probably what is being targeted.

In the absence of such frame designs existing already, this idea would be hard to pitch. The path already seems paved though.

We have e-bikes with bottom bracket motors, which assist the pedals, and drive a chain. We also have bike wheel designs with a hub motor that can retrofit into ordinary bikes.

This looks like it just combines the two: take a bike which has a bottom bracket motor, and replace the motor with a pure generator. Scrap the chain and sprockets, and just deliver electricity to a hub motor.

Chains and sprockets get dirty and require cleaning, except in fully-closed systems that require a complicated transmission. Chains can slip and break. Chains wear out and require replacement, usually together with the rear sprockets. Front rings wear also; about once every three times you change a rear sprocket, you have to change the front rings which, for entry level bikes is usually most cheaply done by getting the entire crank set.

Typically, a multi-speed rear cassette must be removed in order to replace a broken spoke, which is a PITA. This is because it is larger than the hub, and is right next to it, blocking access to the spoke entry holes on the drive side. You need chain whip to prevent the hub from turning, while you apply a wrench to a special lock ring tool, using a great deal of force. From the looks of most wheel builds with a hub motor, it looks like the spokes are easily accessible without removing any difficult part from the wheel: just get the wheel out of the bike.

Internal gears alleviate some of the issues with chains. Without a derailleur system, chains can be encased to protect them from the elements. An electric transmission is going to be more efficient and quieter than internal gears, though, and require no maintenance.

You make some good points, and I'd venture to guess that you're handy with a bike, but for the vast majority of people who use bikes, what you're saying about maintenance is totally irrelevant.

I'm also handy with a bike, and I've spent some time volunteering fixing up used bikes for resale. Here's what I usually saw:

> Chains wear out and require replacement, usually together with the rear sprockets. Front rings wear also; about once every three times you change a rear sprocket, you have to change the front rings which, for entry level bikes is usually most cheaply done by getting the entire crank set.

This is true, but very, very few bikes ever get the kind of miles put on them to wear out a chain. If you ever wear out a chain, you're conservatively in the top 3% of cyclists by miles ridden. Mostly chains die a slow horrible death by being left outside all winter and turning entirely into a rusty immovable mess. If they're completely beyond saving, you cut them off with a hacksaw, put on a new one, check the cassette, and move in to the next neglected thing. I've never replaced a cassette other than on my own bike, no matter how nice or shabby the bike looked.

In the extraordinarily unlikely case that somebody manages to wear out the chain on a cheap bike, they're almost certainly as well off just buying a new cheap bike. Everything on a cheap bike is crap, and by the time you've killed a chain, something else will need to be fixed. Unless the labor is free (mine was, that was the point), you're quickly into more money to polish a turd than to buy a new one.

If by "entry level", you mean "inexpensive from an actual bike shop", then yeah, it might be worth replacing some drivetrain components. Even so, you'd be dealing with the unicorn rider who rides enough to wear out a chain, but also doesn't want to upgrade to a midrange bike.

> Typically, a multi-speed rear cassette must be removed in order to replace a broken spoke, which is a PITA. This is because it is larger than the hub, and is right next to it, blocking access to the spoke entry holes on the drive side. You need chain whip to prevent the hub from turning...

I've broken and replaced spokes. It's pretty rare, and it requires special tools. Most cyclists can't adjust a derailleur, and a surprisingly large number are incapable of fixing a flat.

Everything you're saying is true, but for nearly everybody, a broken spoke is a job for a bike shop. The barrier for replacing a spoke for most people isn't having the tools, it's knowing how to use them, and specifically how to re-true the wheel. I don't mean getting it perfect on a stand, I mean getting it rideable using the brake pads for a reference.

I know somebody who used to race competitively. He has a shop do all his maintenance because he's not even the slightest bit handy, and I can only assume it's not for the lack of opportunity to learn while he was riding competitively.

Maintenance is not "totally irrelevant" to people who have someone else do it.

> This is true, but very, very few bikes ever get the kind of miles put on them to wear out a chain.

I suspect that is changing with e-bikes which are used for actual daily commuting. People buy those things to ride them.

In my experience, daily commuting of around 20 km on a non-powered ordinary bike all year round requires a yearly chain replacement. You can get away with a once per two years cassette job.

Now these e-bikes have serious torque and power. You regularly see them keeping up with cars going 50 km/h or more, even uphill. Yet may use use ordinary drive trains, such as entry-level Shimano cassettes, derailleurs and rings. The power of the motor can easily be expected to trash these components way faster than a human power.

> In the extraordinarily unlikely case that somebody manages to wear out the chain on a cheap bike, they're almost certainly as well off just buying a new cheap bike.

This is false, because chains cost something like $15-$30. Cheap chains and drive sets are still found on entry-level real bikes that might go for $600-$800 or whatever. You're not going to replace an $800 bike because of the chain.

It wouldn't make sense to replace even a $100 bike-shaped-object if all it needs is a $15 chain. There was a time when I rode crap bikes; I still maintained them, and replaced the chain.

> I've broken and replaced spokes. It's pretty rare.

During my daily cycling era when I didn't have a car for some 8 years, I fixed about 3 broken spokes per year. Almost always on the drive side of the rear wheel. So while rarer than a flat tire, it's not that rare.

I had no time for bike shops. They are too far away and have stupid hours like not opening until 10:30 on a weekday, and being closed Sundays. They will keep your bike for at least a day, and charge some ridiculous amount to change a part that costs a dollar.

Truing a wheel is not difficult (particularly lateral-only truing), and once you go to disc brakes, perfection is less important. The bike can even be ridden for a few days with a broken spoke: you just have to clip it off, because you don't want a broken spoke flailing around. It's not a great idea to ride a wheel with a broken spoke, but lets you schedule a good time to fix it, if you're busy

What I'd pay a shop to do would be axial truing: fixing the eccentricity of the wheel's circle. I asume that if I tell a shop to true a wheel, they are only going to care about left-right wobble, which "anyone" can do.

>I fixed about 3 broken spokes per year. Almost always on the drive side of the rear wheel.

This hints at one less obvious advantage of the drive system described in the article: no need for asymmetrical wheel dishing. Rear wheels for conventional derailleur systems have to be built asymmetrically[0]. This makes the wheel weaker than a symmetrical one, because the drive side spokes are under higher tension.

I ride a hub gear bicycle with symmetrical wheels as my primary form of transport and I've never broken a spoke. Electric drive would allow the same wheel strength.

[0] https://en.wikipedia.org/wiki/Bicycle_wheel#Dish

Let me put it in programmer terms: you and I are like the 1% of browser users who open the developer tools. We exist, but to the larger bike industry the number of people like us rounds to zero.

You also fall into one of the (as far as I can tell) two groups that skew heavily towards caring about end-user maintainability: hard-core transportation cyclists (you, and to a much lesser extent me), and people who bike tour (me as well, though not recently)

Thanks to the recent proliferation of e-bikes, driven by battery tech improvements, many more people are looking at bikes as viable urban transportation. People who are relatively unfit, and would never commute on a human-powered bike, day in and day out, are zooming around town on these things, logging many miles. Maintenance issues are going to matter.

People are not going to fork up two thousand dollars just to have the thing sitting in storage; they are looking to it to solve a transportation problem.

Only the spoke-changing maintenance issue applies to hard-core DIY people; the remainder of the point is that the electric transmission likely requires next to zero maintenance. So for the non-DIY people who don't maintain bicycle drive trains themselves, this still means less downtime and expense.

If a spoke is easier to change, that could make a difference to your bike shop visit: "sure, I have some time to do that within the next one to two hours" versus "I can have that ready by noon tomorrow".

It's really an e-bike with a pedal powered generator. But it has a control system to create the illusion that the pedals are stiffly connected to the wheels.

The e-bike people try too hard to pretend they're not building light-duty motorcycles. This seems to be changing. The wheels are getting smaller and stronger, and the center of gravity is going down.

See also bike2.dk as a 'chainless electric bicycle' series hybrid bicycle drivetrain - https://www.youtube.com/watch?v=4JKOKpg21dQ&t=313s (2016).

Like others, it didn't at first make sense to me when thinking about efficiency losses and material complexity, but later realized it might have a place in certain battery-electric delivery bike applications where the majority of the power will be coming from a pre-charged battery. Think about how one could, in a worst off case, charge the bike pedaling in standsill position in a shaded area (while reading a book or using your phone) before trying to continue using the bike.

It also can make a lot of sense if you're assuming ebike, and thus that the material complexity is already 90% there (if not more, a generator is not really a complex beast after all).

And between the limited maintenance and the ebike-engine, direct-drive efficiency also becomes less of a concern.

Would it still be a bicycle if the pedals aren't connected to the drive? This configuration would make it more like a scooter with a pedal generator.

"it’s an electric motorcycle with a foot-operated charging crank."

And you can use it to charge your OLPC XO-1! ;)

Exactly, I am not sure if they would be allowed to be in the bicycle lanes in European countries (or at least in Germany)

Do they not allow e-bikes in those? That seems surprising, given how important e-bikes and pedal-assists have been (or have been represented as being) for broadening the accessibility of cycling.

A quick googling led me to this statement about Germany and e-bikes:

"Insurance and license plates are required. The maximum motor output is 500 watts for e-bikes. Also, e-bike drivers must use bike lanes unless there are none, in which case they are allowed to ride on the roads."

From: https://adoebike.it/en/an-ultimate-guide-to-e-bike-laws-in-e...

E-bikes are generally allowed as they 1) work like regular bikes. 2) only allowed if they're pedal assisted, ie not mopeds.

Of course e bikes are allowed, but as you said you need an insurance + they must be capped to 25kmh and only provide power when the user is actively pedalling.

in NL we have 3 kinds rly. 1) electrical assist that is limited to matching the riders output and limited in speed. They are cosidered bicycles. 2) e-bikes that are considered mopeds and 3) electric motorcycles.

That differs from country to country.

In France, pedelecs (pedal-assist, assist is capped to 25km/h) are allowed on bike lanes, but speed bikes (not limited to 25km/h, they can reach the same speed as cars in cities, often 50km/h, and do not require pedaling) are not allowed on bike lanes.

For all intent and purpose, a speed bike (or however you call it) is an electric moped.

Electric scooters are already allowed (unfortunately for cyclists...), why would adding a charge-crank change that?

I'd be more concerned about how a drive system like that would subtly fail to tickle the endorphine feedback loops the way a bike does. I don't think you'd notice on a conscious level, chances are you might even think you enjoy getting button-press acceleration from the battery buffer, but riding a bike has an immediacy that is close to the walking/running evolution has wired us for. I doubt that an e-scooter fueled by an ergometer generator would come anywhere close. I'm somewhat involved with a cycling vacation business and the way a day of being exposed to those feedback loops makes everybody involved happy that business feels almost like cheating. I really doubt that "ergometer driven e-scooter" could ever come anywhere close to that. But, well, Schaeffler isn't aiming at recreational cycling at all, just at the last mile delivery industry. I could not even guess wether it would make those jobs even more miserable or not.

See these:



Quote from the second article:

The EU Commission’s statement ends an almost 5-year problem that has unsettled and set back manufacturers investing in the development of series hybrid bicycles. Undoubtedly, this technology is still a niche product but manufacturers such as automotive supplier Schaeffler are increasingly investing in this technology.

Ignoring that they're not too different from an e-scooter, these could trivially only power the cycle when the user is actively pedalling, which would be the main issue with respect to pedelec classification.

Though interestingly it could also put a hard limit on upper speed, since "motor assistance" has to cutoff at 25km/h. A series hybrid bike might be considered to only work off of motor assistance (ignoring downhill).

Gonna make a guess that this is not meant to be used as a source of propulsion outside the e-bike ecosystem. To lllustrate, here in Toronto, e-bike must have pedals that can be used to actually move the bike. Electric assist must be "pedal assist" and not throttle based. My guess is that the Free Drive system is intended for that purpose.

> The regenerative solution is a serial hybrid drive that converts the mechanical energy generated during pedaling into electric energy, which in turn is converted back into mechanical energy in the wheel hub motor.

How much efficiency is lost compared to a direct mechanical system?

That's what I wonder too. Honestly, the efficiency of a chain drive system is really high. (however, it should match your body's efficiency - 90 is supposed to be an excellent cadence for cardio, and is good for the knees)

wikipedia says up to 99%:

Mechanical efficiency

From a mechanical viewpoint, up to 99% of the energy delivered by the rider into the pedals is transmitted to the wheels (clean, lubricated new chain at 400 W), although the use of gearing mechanisms reduces this by 1–7% (clean, well-lubricated derailleurs), 4–12% (chain with 3-speed hubs), or 10–20% (shaft drive with 3-speed hubs). The higher efficiencies in each range are achieved at higher power levels and in direct drive (hub gears) or with large driven cogs (derailleurs).


Would be nice to have one of these pedal units installed on a footstool, so that I could charge my phone while watching a video on the couch.

Is anybody selling these in a complete bike yet?

An AWD bicycle would be pretty neat.

AWD motorcycles have been done. They are useful in extreme off-road scenarios.


Does peddling really generate enough watts to fully power the rear wheel at comparable speed/torque level?

I would say that will depend on the rider and the level of pain they want to go through. Some ballpark figures: A pro-level male athlete can comfortable put out 200W without breaking a sweat. All-out effort for an elite rider (male) is upwards of 400W for 20+ min effort. (According to Strava records Van der Poel has put out 450W for 20min) A pro-level female rider can put out 300-350W continuously for an hour (and longer) in all-out effort. A semi-untrained (male) individual can put out ~150-200W on average for an hour of all-out effort. 180W is ~30km/h for an amateur rider in a comfort position on a endurance geometry racing bike with system weight of 80kg. A semi-untrained rider would consider 90-130W acceptable for a leisure ride.

These figures are quite less useless, you've to put them in perspective with the system weight (bike + rider), just can't escape physics.

I don’t see how it couldn’t. I mean, the wheels spin when you pedal on a normal bike, don’t they?

No it's obviously less efficient than a mechanical drive. The point is that it gives more flexibility in design and energy management, and may be more mechanically robust. So the benefits can be worth the loss of efficiency for certain applications.

The applications where additional batteries are not preferable to this complicated system are vanishingly slim IMO. I've seen this proposed over the years over and over, with people designing similar systems, and i've always ended up disappointed in the objective performance characteristics of them.

Fair enough. It’s certainly interesting.

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